{ "schema_version": "kingy-price-index-provider-refresh-v1", "release_id": "kapi-price-index-2026-08-24T041735Z", "generated_at": "2026-08-24T04:17:35.255Z", "publication_authorized": true, "predecessor": { "release_id": "kapi-price-index-2026-08-23T042959Z", "sha256": "4d7d5e5a8a7cea0014ec6d0a2e481fecbafc9da2575e4b47deb35b94311e52d5", "path": "/Users/curtispyke/Documents/Codex/2026-08-22/rev/outputs/ai-api-price-index-rc3-2026-08-22/price-index-release-candidate.json" }, "models": [ { "model_id": "kng:anthropic.claude-fable-5", "provider_id": "kng:provider:anthropic", "model_name": "Claude Fable 5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "api_model_id": "claude-fable-5", "aliases": [ "claude-fable-5" ], "access_route": "first_party_api", "modalities": [ "text" ], "availability_status": "available" }, { "model_id": "kng:anthropic.claude-mythos-5", "provider_id": "kng:provider:anthropic", "model_name": "Claude Mythos 5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "api_model_id": "claude-mythos-5", "aliases": [ "claude-mythos-5" ], "access_route": "first_party_api", "modalities": [ "text" ], "availability_status": "limited" }, { "model_id": "kng:anthropic.claude-opus-5", "provider_id": "kng:provider:anthropic", "model_name": "Claude Opus 5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "api_model_id": "claude-opus-5", "aliases": [ "claude-opus-5" ], "access_route": "first_party_api", "modalities": [ "text" ], "availability_status": "available" }, { "model_id": "kng:anthropic.claude-platform-tools", "provider_id": "kng:provider:anthropic", "model_name": "Claude platform tools", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-platform-tools", "api_model_id": "claude-platform-tools", "aliases": [ "claude-platform-tools" ], "access_route": "first_party_api", "modalities": [ "other" ], "availability_status": "available" }, { "model_id": "kng:anthropic.claude-sonnet-5", "provider_id": "kng:provider:anthropic", "model_name": "Claude Sonnet 5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "api_model_id": "claude-sonnet-5", "aliases": [ "claude-sonnet-5" ], "access_route": "first_party_api", "modalities": [ "text" ], "availability_status": "available" }, { "model_id": "kng:openai.api-tools", "provider_id": "kng:provider:openai", "model_name": "OpenAI API tools", "offering_id": "kapi:offering:openai:first-party-api:openai-api-tools", "api_model_id": "openai-api-tools", "aliases": [ "openai-api-tools" ], "access_route": "first_party_api", "modalities": [ "other" ], "availability_status": "available" }, { "model_id": "kng:openai.gpt-5.6-luna", "provider_id": "kng:provider:openai", "model_name": "GPT-5.6 Luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "api_model_id": "gpt-5.6-luna", "aliases": [ "gpt-5.6-luna" ], "access_route": "first_party_api", "modalities": [ "text" ], "availability_status": "available" }, { "model_id": "kng:openai.gpt-5.6-sol", "provider_id": "kng:provider:openai", "model_name": "GPT-5.6 Sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "api_model_id": "gpt-5.6-sol", "aliases": [ "gpt-5.6-sol" ], "access_route": "first_party_api", "modalities": [ "text" ], "availability_status": "available" }, { "model_id": "kng:openai.gpt-5.6-terra", "provider_id": "kng:provider:openai", "model_name": "GPT-5.6 Terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "api_model_id": "gpt-5.6-terra", "aliases": [ "gpt-5.6-terra" ], "access_route": "first_party_api", "modalities": [ "text" ], "availability_status": "available" } ], "sources": [ { "source_id": "kapi:source:anthropic:batch-2026-08-23", "url": "https://platform.claude.com/docs/en/build-with-claude/batch-processing", "publisher": "Anthropic", "title": "Claude Message Batches guide", "source_type": "official_api_documentation", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "anthropic_first_party_api", "first_captured_at": "2026-08-24T04:11:22.000Z", "last_captured_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "142679774af1409e4422be4de9747ceef3ec251e5c547013e9ae95252fd378a3", "capture_ref": "evidence/anthropic-batch-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: Message Batches | 50%." }, { "source_id": "kapi:source:anthropic:caching-2026-08-23", "url": "https://platform.claude.com/docs/en/build-with-claude/prompt-caching", "publisher": "Anthropic", "title": "Claude prompt caching guide", "source_type": "official_api_documentation", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "anthropic_first_party_api", "first_captured_at": "2026-08-24T04:11:20.000Z", "last_captured_at": "2026-08-24T04:11:20.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "ac27ed4d412018ec84290e8d389c6ae38d5355208a37f7ae5c2ce547217fffb0", "capture_ref": "evidence/anthropic-caching-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: 5-minute | 1-hour | cache_read." }, { "source_id": "kapi:source:anthropic:models-2026-08-23", "url": "https://platform.claude.com/docs/en/about-claude/models/overview", "publisher": "Anthropic", "title": "Claude model overview", "source_type": "official_model_documentation", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "anthropic_first_party_api", "first_captured_at": "2026-08-24T04:11:19.000Z", "last_captured_at": "2026-08-24T04:11:19.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "d2949cf71fa181219472ca31af627f08291aa730336d4c8a3d46563a0c4b5d6b", "capture_ref": "evidence/anthropic-models-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: claude-fable-5 | claude-opus-5 | claude-sonnet-5." }, { "source_id": "kapi:source:anthropic:pricing-2026-08-23", "url": "https://platform.claude.com/docs/en/about-claude/pricing", "publisher": "Anthropic", "title": "Claude API pricing", "source_type": "official_pricing", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "anthropic_first_party_api", "first_captured_at": "2026-08-24T04:11:18.000Z", "last_captured_at": "2026-08-24T04:11:18.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "80682c8d6d33d19e0f7f891385d9b5977dc4f6f025c6618c4e1d6ef2c3e5d04e", "capture_ref": "evidence/anthropic-pricing-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: Claude Fable 5 | Claude Opus 5 | Claude Sonnet 5 | Batch processing." }, { "source_id": "kapi:source:openai:batch-2026-08-23", "url": "https://developers.openai.com/api/docs/guides/batch", "publisher": "OpenAI", "title": "OpenAI Batch guide", "source_type": "official_api_documentation", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "openai_first_party_api", "first_captured_at": "2026-08-24T04:11:14.000Z", "last_captured_at": "2026-08-24T04:11:14.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "07e237f99672b136c391f084d25aa4da23a6647be747d00db70764eb69b372a0", "capture_ref": "evidence/openai-batch-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: 50% | 24 hours." }, { "source_id": "kapi:source:openai:caching-2026-08-23", "url": "https://developers.openai.com/api/docs/guides/prompt-caching", "publisher": "OpenAI", "title": "OpenAI prompt caching guide", "source_type": "official_api_documentation", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "openai_first_party_api", "first_captured_at": "2026-08-24T04:11:17.000Z", "last_captured_at": "2026-08-24T04:11:17.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "a0c1641daaf5d5be9eb7fe95798391a7b75f2c6342d844518cd2bbe56d2a4e83", "capture_ref": "evidence/openai-caching-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: Prompt caching | cached." }, { "source_id": "kapi:source:openai:fast-2026-08-23", "url": "https://developers.openai.com/api/docs/guides/fast-mode", "publisher": "OpenAI", "title": "OpenAI Fast mode guide", "source_type": "official_api_documentation", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "openai_first_party_api", "first_captured_at": "2026-08-24T04:11:16.000Z", "last_captured_at": "2026-08-24T04:11:16.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "d026c73a51089a6ee8e6f45b990a5618224f4a778b6d82176cf5a4786d49b50e", "capture_ref": "evidence/openai-fast-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: Fast mode | Priority processing." }, { "source_id": "kapi:source:openai:flex-2026-08-23", "url": "https://developers.openai.com/api/docs/guides/flex-processing", "publisher": "OpenAI", "title": "OpenAI Flex processing guide", "source_type": "official_api_documentation", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "openai_first_party_api", "first_captured_at": "2026-08-24T04:11:15.000Z", "last_captured_at": "2026-08-24T04:11:15.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "a4625caa136f17c61638d452cbd8ffa6157bab8c1da38f4d63f27bf13bdc9a73", "capture_ref": "evidence/openai-flex-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: Flex processing | lower costs." }, { "source_id": "kapi:source:openai:models-2026-08-23", "url": "https://developers.openai.com/api/docs/models", "publisher": "OpenAI", "title": "OpenAI model catalog", "source_type": "official_model_documentation", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "openai_first_party_api", "first_captured_at": "2026-08-24T04:11:13.000Z", "last_captured_at": "2026-08-24T04:11:13.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "6efa7a1dbcbccba6e087d880ccd47ac17af2fd1dab1f33da1a0b4cc23bec9a2f", "capture_ref": "evidence/openai-models-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: GPT-5.6 Sol | GPT-5.6 Terra | GPT-5.6 Luna." }, { "source_id": "kapi:source:openai:pricing-2026-08-23", "url": "https://developers.openai.com/api/docs/pricing", "publisher": "OpenAI", "title": "OpenAI API pricing", "source_type": "official_pricing", "authority": "vendor_documentation", "language": "en-US", "geographic_scope": "openai_first_party_api", "first_captured_at": "2026-08-24T04:11:12.000Z", "last_captured_at": "2026-08-24T04:11:12.000Z", "announced_at": null, "effective_at": null, "http_status": 200, "content_sha256": "abec04012cc426cb2f75ec4b8db305b8b09ad2965f7be0c446b316bbb698b8b6", "capture_ref": "evidence/openai-pricing-2026-08-23/source.body", "review_state": "approved_public", "notes": "Immutable local capture. Required terms passed: gpt-5.6-sol | gpt-5.6-terra | gpt-5.6-luna | Fast mode." } ], "price_records": [ { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "20 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "6.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "6.25", "normalization_formula": "12.5 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.5", "normalization_formula": "1 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "10 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-batch-50-percent-batch-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "25", "normalization_formula": "50 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "20", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "20", "normalization_formula": "20 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "12.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "12.5", "normalization_formula": "12.5 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1", "normalization_formula": "1 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "10 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-global-routing-supported-countries-standard-on-demand-on-demand-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "50", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "50", "normalization_formula": "50 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "11", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "11", "normalization_formula": "20 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "6.875", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "6.875", "normalization_formula": "12.5 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.55", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.55", "normalization_formula": "1 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5.5", "normalization_formula": "10 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-batch-50-percent-1-1x-us-batch-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "27.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "27.5", "normalization_formula": "50 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "22", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "22", "normalization_formula": "20 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "13.75", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "13.75", "normalization_formula": "12.5 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "1.1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.1", "normalization_formula": "1 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "11", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "11", "normalization_formula": "10 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-fable-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-fable-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-fable-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "55", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "55", "normalization_formula": "50 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "20 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "6.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "6.25", "normalization_formula": "12.5 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.5", "normalization_formula": "1 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "10 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-batch-50-percent-batch-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "25", "normalization_formula": "50 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "20", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "20", "normalization_formula": "20 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "12.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "12.5", "normalization_formula": "12.5 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1", "normalization_formula": "1 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "10 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-global-routing-supported-countries-standard-on-demand-on-demand-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "50", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "50", "normalization_formula": "50 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "11", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "11", "normalization_formula": "20 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "6.875", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "6.875", "normalization_formula": "12.5 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.55", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.55", "normalization_formula": "1 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5.5", "normalization_formula": "10 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-batch-50-percent-1-1x-us-batch-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "27.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "27.5", "normalization_formula": "50 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "22", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "22", "normalization_formula": "20 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "13.75", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "13.75", "normalization_formula": "12.5 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "1.1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.1", "normalization_formula": "1 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "11", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "11", "normalization_formula": "10 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-mythos-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-mythos-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-mythos-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "55", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "55", "normalization_formula": "50 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "10 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "3.125", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "3.125", "normalization_formula": "6.25 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.25", "normalization_formula": "0.5 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "2.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.5", "normalization_formula": "5 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-batch-50-percent-batch-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "12.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "12.5", "normalization_formula": "25 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-fast-direct-api-priority-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-fast-direct-api-priority-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "fast_direct_api", "billing_mode": "priority", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "native_amount (USD per 1M tokens; no FX conversion)", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Fast mode is limited to the direct Claude API and is not available with Message Batches. Current published Fast pricing is $10 input / $50 output per 1M tokens." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-fast-direct-api-priority-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-fast-direct-api-priority-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "fast_direct_api", "billing_mode": "priority", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "50", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "50", "normalization_formula": "native_amount (USD per 1M tokens; no FX conversion)", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Fast mode is limited to the direct Claude API and is not available with Message Batches. Current published Fast pricing is $10 input / $50 output per 1M tokens." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "10 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "6.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "6.25", "normalization_formula": "6.25 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.5", "normalization_formula": "0.5 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "5 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-global-routing-supported-countries-standard-on-demand-on-demand-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "25", "normalization_formula": "25 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5.5", "normalization_formula": "10 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "3.4375", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "3.4375", "normalization_formula": "6.25 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.275", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.275", "normalization_formula": "0.5 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "2.75", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.75", "normalization_formula": "5 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-batch-50-percent-1-1x-us-batch-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "13.75", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "13.75", "normalization_formula": "25 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "11", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "11", "normalization_formula": "10 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "6.875", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "6.875", "normalization_formula": "6.25 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.55", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.55", "normalization_formula": "0.5 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5.5", "normalization_formula": "5 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-opus-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-opus-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-opus-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "27.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "27.5", "normalization_formula": "25 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-platform-tools-global-routing-supported-countries-code-execution-on-demand-tool-call-na-na", "claim_id": "kapi:anthropic:claim:claude-platform-tools-global-routing-supported-countries-code-execution-on-demand-tool-call-na-na", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-platform-tools", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-platform-tools", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "code_execution", "billing_mode": "on_demand", "modality": "other", "price_component": "tool_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0.05", "currency": "USD", "native_unit": "per_hour", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Standalone code execution has a 5-minute minimum; the first 1,550 organization hours each month are free, then $0.05 per container-hour. No separate compute charge when paired with web search or web fetch, excluding token charges." }, { "record_id": "kapi:price-record:anthropic:claude-platform-tools-global-routing-supported-countries-web-fetch-on-demand-tool-call-na-na", "claim_id": "kapi:anthropic:claim:claude-platform-tools-global-routing-supported-countries-web-fetch-on-demand-tool-call-na-na", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-platform-tools", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-platform-tools", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "web_fetch", "billing_mode": "on_demand", "modality": "other", "price_component": "tool_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0", "currency": "USD", "native_unit": "per_call", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Web fetch has no per-call fee, but fetched content is billed as input tokens." }, { "record_id": "kapi:price-record:anthropic:claude-platform-tools-global-routing-supported-countries-web-search-on-demand-search-call-na-na", "claim_id": "kapi:anthropic:claim:claude-platform-tools-global-routing-supported-countries-web-search-on-demand-search-call-na-na", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-platform-tools", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-platform-tools", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "web_search", "billing_mode": "on_demand", "modality": "other", "price_component": "search_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0.01", "currency": "USD", "native_unit": "per_call", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Web search costs $10 per 1,000 searches; tokens used by search are billed separately." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2", "normalization_formula": "4 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "1.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.25", "normalization_formula": "2.5 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.1", "normalization_formula": "0.2 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1", "normalization_formula": "2 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-batch-50-percent-batch-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "batch_50_percent", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "10 × 0.5 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4", "normalization_formula": "4 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "2.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.5", "normalization_formula": "2.5 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.2 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2", "normalization_formula": "2 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-global-routing-supported-countries-standard-on-demand-on-demand-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "global_routing_supported_countries", "service_tier": "standard_on_demand", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "10 × 1 batch factor × 1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Global inference is the direct API default. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "2.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.2", "normalization_formula": "4 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "1.375", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.375", "normalization_formula": "2.5 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.11", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.11", "normalization_formula": "0.2 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "1.1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.1", "normalization_formula": "2 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-batch-50-percent-1-1x-us-batch-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "batch_50_percent_1_1x_us", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "5.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5.5", "normalization_formula": "10 × 0.5 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23", "kapi:source:anthropic:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-1h-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-1h-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_1h", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "4.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4.4", "normalization_formula": "4 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-5m-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cache-write-5m-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write_5m", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "2.75", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.75", "normalization_formula": "2.5 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cached-input-read-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-cached-input-read-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "0.22", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.22", "normalization_formula": "0.2 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-input-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-input-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "2.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.2", "normalization_formula": "2 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:anthropic:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-output-0-1000000", "claim_id": "kapi:anthropic:claim:claude-sonnet-5-us-only-inference-standard-on-demand-1-1x-us-on-demand-output-0-1000000", "revision": 1, "provider_id": "kng:provider:anthropic", "model_id": "kng:anthropic.claude-sonnet-5", "offering_id": "kapi:offering:anthropic:first-party-api:anthropic-claude-sonnet-5", "access_route": "first_party_api", "region": "us_only_inference", "service_tier": "standard_on_demand_1_1x_us", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 1000000, "context_rule": "whole_request", "native_amount": "11", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "11", "normalization_formula": "10 × 1 batch factor × 1.1 inference-geography factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:anthropic:pricing-2026-08-23", "kapi:source:anthropic:models-2026-08-23", "kapi:source:anthropic:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Claude 5 supports up to 1M input context at the same listed token rate. Direct API US-only inference uses the documented 1.1× multiplier. Claude 4.7+ uses a newer tokenizer that may produce about 30% more tokens for the same text; actual change is workload dependent." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-batch-50-percent-long-context-batch-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-batch-50-percent-long-context-batch-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.25", "normalization_formula": "0.25 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-batch-50-percent-long-context-batch-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-batch-50-percent-long-context-batch-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.02", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.02", "normalization_formula": "0.02 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-batch-50-percent-long-context-batch-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-batch-50-percent-long-context-batch-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.2 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-batch-50-percent-long-context-batch-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-batch-50-percent-long-context-batch-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.9", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.9", "normalization_formula": "1.2 × 0.5 batch_50_percent × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-batch-50-percent-short-context-batch-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-batch-50-percent-short-context-batch-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.125", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.125", "normalization_formula": "0.25 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-batch-50-percent-short-context-batch-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-batch-50-percent-short-context-batch-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.01", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.01", "normalization_formula": "0.02 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-batch-50-percent-short-context-batch-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-batch-50-percent-short-context-batch-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.1", "normalization_formula": "0.2 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-batch-50-percent-short-context-batch-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-batch-50-percent-short-context-batch-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.6", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.6", "normalization_formula": "1.2 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-fast-2x-standard-long-context-priority-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-fast-2x-standard-long-context-priority-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1", "normalization_formula": "0.25 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-fast-2x-standard-long-context-priority-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-fast-2x-standard-long-context-priority-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.08", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.08", "normalization_formula": "0.02 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-fast-2x-standard-long-context-priority-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-fast-2x-standard-long-context-priority-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.8", "normalization_formula": "0.2 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-fast-2x-standard-long-context-priority-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-fast-2x-standard-long-context-priority-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "3.6", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "3.6", "normalization_formula": "1.2 × 2 fast_2x_standard × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-fast-2x-standard-short-context-priority-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-fast-2x-standard-short-context-priority-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.5", "normalization_formula": "0.25 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-fast-2x-standard-short-context-priority-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-fast-2x-standard-short-context-priority-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.04", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.04", "normalization_formula": "0.02 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-fast-2x-standard-short-context-priority-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-fast-2x-standard-short-context-priority-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.4", "normalization_formula": "0.2 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-fast-2x-standard-short-context-priority-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-fast-2x-standard-short-context-priority-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.4", "normalization_formula": "1.2 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-flex-50-percent-long-context-flex-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-flex-50-percent-long-context-flex-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.25", "normalization_formula": "0.25 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-flex-50-percent-long-context-flex-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-flex-50-percent-long-context-flex-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.02", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.02", "normalization_formula": "0.02 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-flex-50-percent-long-context-flex-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-flex-50-percent-long-context-flex-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.2 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-flex-50-percent-long-context-flex-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-flex-50-percent-long-context-flex-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.9", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.9", "normalization_formula": "1.2 × 0.5 flex_50_percent × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-flex-50-percent-short-context-flex-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-flex-50-percent-short-context-flex-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.125", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.125", "normalization_formula": "0.25 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-flex-50-percent-short-context-flex-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-flex-50-percent-short-context-flex-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.01", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.01", "normalization_formula": "0.02 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-flex-50-percent-short-context-flex-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-flex-50-percent-short-context-flex-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.1", "normalization_formula": "0.2 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-flex-50-percent-short-context-flex-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-flex-50-percent-short-context-flex-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.6", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.6", "normalization_formula": "1.2 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.5", "normalization_formula": "0.25 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.04", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.04", "normalization_formula": "0.02 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-standard-on-demand-long-context-on-demand-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-standard-on-demand-long-context-on-demand-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.4", "normalization_formula": "0.2 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-standard-on-demand-long-context-on-demand-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-standard-on-demand-long-context-on-demand-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "1.8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.8", "normalization_formula": "1.2 × 1 standard_on_demand × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-standard-on-demand-short-context-on-demand-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-standard-on-demand-short-context-on-demand-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.25", "normalization_formula": "0.25 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.02", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.02", "normalization_formula": "0.02 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-standard-on-demand-short-context-on-demand-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-standard-on-demand-short-context-on-demand-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.2 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-global-default-processing-standard-on-demand-short-context-on-demand-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-global-default-processing-standard-on-demand-short-context-on-demand-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "1.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.2", "normalization_formula": "1.2 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-luna. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.55", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.55", "normalization_formula": "0.25 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.044", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.044", "normalization_formula": "0.02 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.44", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.44", "normalization_formula": "0.2 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "1.98", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.98", "normalization_formula": "1.2 × 1.5 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.275", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.275", "normalization_formula": "0.25 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.022", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.022", "normalization_formula": "0.02 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.22", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.22", "normalization_formula": "0.2 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-luna-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-luna", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-luna", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "1.32", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.32", "normalization_formula": "1.2 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-batch-50-percent-long-context-batch-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-batch-50-percent-long-context-batch-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "5 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-batch-50-percent-long-context-batch-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-batch-50-percent-long-context-batch-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.4", "normalization_formula": "0.4 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-batch-50-percent-long-context-batch-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-batch-50-percent-long-context-batch-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4", "normalization_formula": "4 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-batch-50-percent-long-context-batch-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-batch-50-percent-long-context-batch-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "15", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "15", "normalization_formula": "20 × 0.5 batch_50_percent × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-batch-50-percent-short-context-batch-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-batch-50-percent-short-context-batch-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.5", "normalization_formula": "5 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-batch-50-percent-short-context-batch-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-batch-50-percent-short-context-batch-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.4 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-batch-50-percent-short-context-batch-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-batch-50-percent-short-context-batch-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2", "normalization_formula": "4 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-batch-50-percent-short-context-batch-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-batch-50-percent-short-context-batch-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "20 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-fast-2x-standard-long-context-priority-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-fast-2x-standard-long-context-priority-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "20", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "20", "normalization_formula": "5 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-fast-2x-standard-long-context-priority-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-fast-2x-standard-long-context-priority-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "1.6", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.6", "normalization_formula": "0.4 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-fast-2x-standard-long-context-priority-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-fast-2x-standard-long-context-priority-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "16", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "16", "normalization_formula": "4 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-fast-2x-standard-long-context-priority-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-fast-2x-standard-long-context-priority-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "60", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "60", "normalization_formula": "20 × 2 fast_2x_standard × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-fast-2x-standard-short-context-priority-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-fast-2x-standard-short-context-priority-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "5 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-fast-2x-standard-short-context-priority-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-fast-2x-standard-short-context-priority-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.8", "normalization_formula": "0.4 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-fast-2x-standard-short-context-priority-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-fast-2x-standard-short-context-priority-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "8", "normalization_formula": "4 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-fast-2x-standard-short-context-priority-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-fast-2x-standard-short-context-priority-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "40", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "40", "normalization_formula": "20 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-flex-50-percent-long-context-flex-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-flex-50-percent-long-context-flex-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "5 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-flex-50-percent-long-context-flex-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-flex-50-percent-long-context-flex-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.4", "normalization_formula": "0.4 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-flex-50-percent-long-context-flex-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-flex-50-percent-long-context-flex-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4", "normalization_formula": "4 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-flex-50-percent-long-context-flex-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-flex-50-percent-long-context-flex-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "15", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "15", "normalization_formula": "20 × 0.5 flex_50_percent × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-flex-50-percent-short-context-flex-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-flex-50-percent-short-context-flex-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.5", "normalization_formula": "5 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-flex-50-percent-short-context-flex-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-flex-50-percent-short-context-flex-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.4 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-flex-50-percent-short-context-flex-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-flex-50-percent-short-context-flex-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2", "normalization_formula": "4 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-flex-50-percent-short-context-flex-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-flex-50-percent-short-context-flex-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "20 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "5 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.8", "normalization_formula": "0.4 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-standard-on-demand-long-context-on-demand-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-standard-on-demand-long-context-on-demand-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "8", "normalization_formula": "4 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-standard-on-demand-long-context-on-demand-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-standard-on-demand-long-context-on-demand-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "30", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "30", "normalization_formula": "20 × 1 standard_on_demand × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-standard-on-demand-short-context-on-demand-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-standard-on-demand-short-context-on-demand-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "5 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.4", "normalization_formula": "0.4 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-standard-on-demand-short-context-on-demand-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-standard-on-demand-short-context-on-demand-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4", "normalization_formula": "4 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-global-default-processing-standard-on-demand-short-context-on-demand-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-global-default-processing-standard-on-demand-short-context-on-demand-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "20", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "20", "normalization_formula": "20 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-sol. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "11", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "11", "normalization_formula": "5 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.88", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.88", "normalization_formula": "0.4 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "8.8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "8.8", "normalization_formula": "4 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "33", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "33", "normalization_formula": "20 × 1.5 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "5.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5.5", "normalization_formula": "5 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.44", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.44", "normalization_formula": "0.4 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "4.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4.4", "normalization_formula": "4 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-sol-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-sol", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-sol", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "22", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "22", "normalization_formula": "20 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-batch-50-percent-long-context-batch-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-batch-50-percent-long-context-batch-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "2.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.5", "normalization_formula": "2.5 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-batch-50-percent-long-context-batch-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-batch-50-percent-long-context-batch-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.2 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-batch-50-percent-long-context-batch-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-batch-50-percent-long-context-batch-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2", "normalization_formula": "2 × 0.5 batch_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-batch-50-percent-long-context-batch-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-batch-50-percent-long-context-batch-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_long_context", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "9", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "9", "normalization_formula": "12 × 0.5 batch_50_percent × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-batch-50-percent-short-context-batch-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-batch-50-percent-short-context-batch-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "1.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.25", "normalization_formula": "2.5 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-batch-50-percent-short-context-batch-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-batch-50-percent-short-context-batch-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.1", "normalization_formula": "0.2 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-batch-50-percent-short-context-batch-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-batch-50-percent-short-context-batch-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1", "normalization_formula": "2 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-batch-50-percent-short-context-batch-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-batch-50-percent-short-context-batch-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "batch_50_percent_short_context", "billing_mode": "batch", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "6", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "6", "normalization_formula": "12 × 0.5 batch_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:batch-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-fast-2x-standard-long-context-priority-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-fast-2x-standard-long-context-priority-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "10", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "10", "normalization_formula": "2.5 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-fast-2x-standard-long-context-priority-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-fast-2x-standard-long-context-priority-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.8", "normalization_formula": "0.2 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-fast-2x-standard-long-context-priority-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-fast-2x-standard-long-context-priority-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "8", "normalization_formula": "2 × 2 fast_2x_standard × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-fast-2x-standard-long-context-priority-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-fast-2x-standard-long-context-priority-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_long_context", "billing_mode": "priority", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "36", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "36", "normalization_formula": "12 × 2 fast_2x_standard × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-fast-2x-standard-short-context-priority-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-fast-2x-standard-short-context-priority-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "2.5 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-fast-2x-standard-short-context-priority-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-fast-2x-standard-short-context-priority-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.4", "normalization_formula": "0.2 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-fast-2x-standard-short-context-priority-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-fast-2x-standard-short-context-priority-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4", "normalization_formula": "2 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-fast-2x-standard-short-context-priority-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-fast-2x-standard-short-context-priority-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "fast_2x_standard_short_context", "billing_mode": "priority", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "24", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "24", "normalization_formula": "12 × 2 fast_2x_standard × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:fast-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-flex-50-percent-long-context-flex-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-flex-50-percent-long-context-flex-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "2.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.5", "normalization_formula": "2.5 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-flex-50-percent-long-context-flex-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-flex-50-percent-long-context-flex-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.2 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-flex-50-percent-long-context-flex-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-flex-50-percent-long-context-flex-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2", "normalization_formula": "2 × 0.5 flex_50_percent × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-flex-50-percent-long-context-flex-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-flex-50-percent-long-context-flex-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_long_context", "billing_mode": "flex", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "9", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "9", "normalization_formula": "12 × 0.5 flex_50_percent × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-flex-50-percent-short-context-flex-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-flex-50-percent-short-context-flex-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "1.25", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1.25", "normalization_formula": "2.5 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-flex-50-percent-short-context-flex-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-flex-50-percent-short-context-flex-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.1", "normalization_formula": "0.2 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-flex-50-percent-short-context-flex-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-flex-50-percent-short-context-flex-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "1", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "1", "normalization_formula": "2 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-flex-50-percent-short-context-flex-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-flex-50-percent-short-context-flex-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "flex_50_percent_short_context", "billing_mode": "flex", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "6", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "6", "normalization_formula": "12 × 0.5 flex_50_percent × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23", "kapi:source:openai:flex-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5", "normalization_formula": "2.5 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.4", "normalization_formula": "0.2 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-standard-on-demand-long-context-on-demand-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-standard-on-demand-long-context-on-demand-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4", "normalization_formula": "2 × 1 standard_on_demand × 2 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-standard-on-demand-long-context-on-demand-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-standard-on-demand-long-context-on-demand-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "18", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "18", "normalization_formula": "12 × 1 standard_on_demand × 1.5 long-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. When input exceeds 272K tokens, the long-context rate applies to the entire request: 2× input-side and 1.5× output." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-standard-on-demand-short-context-on-demand-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-standard-on-demand-short-context-on-demand-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.5", "normalization_formula": "2.5 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.2", "normalization_formula": "0.2 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-standard-on-demand-short-context-on-demand-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-standard-on-demand-short-context-on-demand-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2", "normalization_formula": "2 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-global-default-processing-standard-on-demand-short-context-on-demand-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-global-default-processing-standard-on-demand-short-context-on-demand-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "12", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "12", "normalization_formula": "12 × 1 standard_on_demand × 1 short-context factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "POST /v1/responses with gpt-5.6-terra. Short-context band covers requests with at most 272K input tokens." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cache-write-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "5.5", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "5.5", "normalization_formula": "2.5 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-cached-input-read-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "0.44", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.44", "normalization_formula": "0.2 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-input-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-input-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "4.4", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "4.4", "normalization_formula": "2 × 2 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-output-272001-1050000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-long-context-on-demand-output-272001-1050000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_long_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 272001, "context_max_tokens": 1050000, "context_rule": "whole_request", "native_amount": "19.8", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "19.8", "normalization_formula": "12 × 1.5 long-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cache-write-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cache-write-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cache_write", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2.75", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.75", "normalization_formula": "2.5 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-cached-input-read-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "cached_input_read", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "0.22", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "0.22", "normalization_formula": "0.2 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-input-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-input-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "input", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "2.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "2.2", "normalization_formula": "2 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-output-0-272000", "claim_id": "kapi:openai:claim:gpt-5-6-terra-regional-processing-10-percent-uplift-standard-on-demand-short-context-on-demand-output-0-272000", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.gpt-5.6-terra", "offering_id": "kapi:offering:openai:first-party-api:openai-gpt-5-6-terra", "access_route": "first_party_api", "region": "regional_processing_10_percent_uplift", "service_tier": "standard_on_demand_short_context", "billing_mode": "on_demand", "modality": "text", "price_component": "output", "context_min_tokens": 0, "context_max_tokens": 272000, "context_rule": "whole_request", "native_amount": "13.2", "currency": "USD", "native_unit": "per_1m_tokens", "normalized_usd_per_1m_tokens": "13.2", "normalization_formula": "12 × 1 short-context factor × 1.1 regional-processing factor", "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23", "kapi:source:openai:models-2026-08-23", "kapi:source:openai:caching-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Regional processing is encoded separately from global default processing. This row does not project availability of combinations with Batch, Flex, or Fast." }, { "record_id": "kapi:price-record:openai:openai-api-tools-global-default-processing-container-16gb-on-demand-tool-call-na-na", "claim_id": "kapi:openai:claim:openai-api-tools-global-default-processing-container-16gb-on-demand-tool-call-na-na", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.api-tools", "offering_id": "kapi:offering:openai:first-party-api:openai-api-tools", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "container_16gb", "billing_mode": "on_demand", "modality": "other", "price_component": "tool_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0.48", "currency": "USD", "native_unit": "per_request", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "16 GB container price per 20-minute session; 5-minute minimum billing window." }, { "record_id": "kapi:price-record:openai:openai-api-tools-global-default-processing-container-1gb-on-demand-tool-call-na-na", "claim_id": "kapi:openai:claim:openai-api-tools-global-default-processing-container-1gb-on-demand-tool-call-na-na", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.api-tools", "offering_id": "kapi:offering:openai:first-party-api:openai-api-tools", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "container_1gb", "billing_mode": "on_demand", "modality": "other", "price_component": "tool_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0.03", "currency": "USD", "native_unit": "per_request", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "1 GB container price per 20-minute session; 5-minute minimum billing window." }, { "record_id": "kapi:price-record:openai:openai-api-tools-global-default-processing-container-4gb-on-demand-tool-call-na-na", "claim_id": "kapi:openai:claim:openai-api-tools-global-default-processing-container-4gb-on-demand-tool-call-na-na", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.api-tools", "offering_id": "kapi:offering:openai:first-party-api:openai-api-tools", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "container_4gb", "billing_mode": "on_demand", "modality": "other", "price_component": "tool_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0.12", "currency": "USD", "native_unit": "per_request", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "4 GB container price per 20-minute session; 5-minute minimum billing window." }, { "record_id": "kapi:price-record:openai:openai-api-tools-global-default-processing-container-64gb-on-demand-tool-call-na-na", "claim_id": "kapi:openai:claim:openai-api-tools-global-default-processing-container-64gb-on-demand-tool-call-na-na", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.api-tools", "offering_id": "kapi:offering:openai:first-party-api:openai-api-tools", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "container_64gb", "billing_mode": "on_demand", "modality": "other", "price_component": "tool_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "1.92", "currency": "USD", "native_unit": "per_request", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "64 GB container price per 20-minute session; 5-minute minimum billing window." }, { "record_id": "kapi:price-record:openai:openai-api-tools-global-default-processing-file-search-on-demand-tool-call-na-na", "claim_id": "kapi:openai:claim:openai-api-tools-global-default-processing-file-search-on-demand-tool-call-na-na", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.api-tools", "offering_id": "kapi:offering:openai:first-party-api:openai-api-tools", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "file_search", "billing_mode": "on_demand", "modality": "other", "price_component": "tool_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0.0025", "currency": "USD", "native_unit": "per_call", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "File search tool calls cost $2.50 per 1,000 calls." }, { "record_id": "kapi:price-record:openai:openai-api-tools-global-default-processing-file-search-storage-on-demand-storage-na-na", "claim_id": "kapi:openai:claim:openai-api-tools-global-default-processing-file-search-storage-on-demand-storage-na-na", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.api-tools", "offering_id": "kapi:offering:openai:first-party-api:openai-api-tools", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "file_search_storage", "billing_mode": "on_demand", "modality": "other", "price_component": "storage", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0.1", "currency": "USD", "native_unit": "other", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "File search storage costs $0.10 per GB per day after the first 1 GB; native_unit is other because the schema has no per-GB-day unit." }, { "record_id": "kapi:price-record:openai:openai-api-tools-global-default-processing-web-search-on-demand-search-call-na-na", "claim_id": "kapi:openai:claim:openai-api-tools-global-default-processing-web-search-on-demand-search-call-na-na", "revision": 1, "provider_id": "kng:provider:openai", "model_id": "kng:openai.api-tools", "offering_id": "kapi:offering:openai:first-party-api:openai-api-tools", "access_route": "first_party_api", "region": "global_default_processing", "service_tier": "web_search", "billing_mode": "on_demand", "modality": "other", "price_component": "search_call", "context_min_tokens": null, "context_max_tokens": null, "context_rule": "not_applicable", "native_amount": "0.01", "currency": "USD", "native_unit": "per_call", "normalized_usd_per_1m_tokens": null, "normalization_formula": null, "fx_rate_id": null, "observed_at": "2026-08-24T04:11:22.000Z", "announced_at": null, "effective_from": null, "effective_to": null, "is_current": true, "source_ids": [ "kapi:source:openai:pricing-2026-08-23" ], "evidence_level": "official_source", "verification_method": "document_review", "review_state": "approved_public", "freshness": "current", "supersedes_record_id": null, "notes": "Web search costs $10 per 1,000 calls; search-content tokens are billed separately at model token rates." } ] }