AI Model Profile
Qwen4-Max
Qwen4-Max is Alibaba Cloud Model Studio’s hosted multimodal model. The production API ID is qwen4-max, distinct from qwen4-max-preview. It accepts text, images and video and returns text. This profile covers the Singapore / International API route.
- Family
- Qwen4
- Release date
- 2026-11-16
- Status
- Current
- Context window
- 1,000,000 tokens
- Output limit
- Non-thinking: up to 131,072 output tokens; thinking: up to 262,144 combined reasoning and answer tokens, within the total context
- API
- yes
- Open weights
- Unknown
- Local/self-hosted
- Unknown
- Pricing
- Singapore / International Standard API: US$1.50 per million input tokens and US$4.50 per million output tokens, for input up to 1,000,000 tokens. Thinking output charges include reasoning and final answer tokens.
- Evidence state
- Source-verified
Benchmark Caveat
Provider capability descriptions are documentation claims, not measured task quality, speed or reliability. Do not transfer results from qwen4-max-preview or older Qwen generations to qwen4-max.
No new Kingy benchmark or paid API call was run for this model profile. Published guide coverage and any retained tests keep their own dates and scope.
Best for
Evaluating a hosted API for text, image or video understanding and tool-based workflows, after testing the exact task.
Skip if
You require verified downloadable weights, a demonstrated local hardware fit, batch inference on this route, or an established quality guarantee for your own workload.
Strengths
The official API documentation supports text, image and video input, text output, function calling, structured JSON output and a 1,000,000-token total context.
Weaknesses
No local weight artifact, local speed measurement or independent quality ranking is supplied by this profile. Account access and request behavior remain untested.
Agent suitability
Documentation lists function calling and parallel tool calls, structured output, and web search. Supported protocols include OpenAI Chat Completions and Responses, Anthropic Messages and DashScope. Confirm exact route restrictions before integration.
Kingy AI Product Facts
Qwen4-Max
Current statusGenerally available
- Company
- Alibaba Cloud
- Primary job
- Hosted text generation and image/video understanding
- Audience
- Not yet reviewed
- API
- Available
- Evidence
- Source-backed; coverage expanding
- Latest source check
- November 16, 2026
See all tracked product factsPricing, platforms, dependencies, data claims, regions, and timeline
Sources and checks
- Coverage
- 12 of 88 tracked fields currently verified
- Sources
- 3
Source-check dates apply to the tracked facts. Editorial updates elsewhere on this page may have a different date.
Plans and pricing units
- Qwen4-Max Singapore / International Standard input 1.5 USD per million input tokens Standard thinking or non-thinking; 0 < input tokens <= 1,000,000. · Output includes reasoning plus answer. · No cache discount, free quota or Batch discount applied.
- Qwen4-Max Singapore / International Standard output 4.5 USD per million output tokens Standard thinking or non-thinking; 0 < input tokens <= 1,000,000. · Output includes reasoning plus answer. · No cache discount, free quota or Batch discount applied.
Where it runs
- Availability
- Not yet reviewed
Developer access
- API access
- AvailableSource: Qwen4-Max official model documentation
- API documentation
- https://www.alibabacloud.com/help/en/model-studio/qwen4-maxSource: Qwen4-Max official model documentation
Stack and integrations
- Model / provider dependencies
- Not yet reviewed
- Integrations
- Not yet reviewed
Data and regions
- Vendor data-use claim
- Not yet reviewed
- Vendor retention claim
- Not yet reviewed
- Regions
- Available: Singapore / International API route
Launch and latest material update
- Launch
- November 16, 2026
- Latest material update
- Not yet reviewed
Reasoning notes
Non-thinking input ceiling: 991,808 tokens. Thinking input ceiling: 983,616 tokens. Output ceilings are separate upper bounds; input plus generated tokens must still fit the 1,000,000-token context.
API pricing notes
Uncached Standard text-token rates checked November 16, 2026. Cache reads: US$0.075/M; 5-minute writes: US$1.875/M; 1-hour writes: US$3/M. Batch inference is not supported on this route. Other regions and providers have separate tariffs. Free-token offers depend on eligibility and are excluded from estimates.
License notes
Hosted Alibaba Cloud API terms apply to this route. No open-weight or open-source license is established by the API documentation reviewed here.
Hardware requirements
This record covers a hosted API. No official downloadable Qwen4-Max weight file or self-hosted hardware requirements were verified in this review. It is not a Local AI hardware recommendation.
Official Model Links
Model Intelligence Research Map
Use these internal paths to move from this model profile into provider pages, static comparison pages, related Kingy records, and the broader AI launch graph. These are research paths, not rankings.