{
  "kind": "observed-local-model-run",
  "packageVersion": "20261029.2",
  "observedAt": "2026-10-30T02:36:18.243173+00:00",
  "passed": false,
  "attempts": 1,
  "modelRequestAttempts": 1,
  "modelResponseReceived": true,
  "inferencePerformed": true,
  "testMethodsRun": 7,
  "failedMethods": [
    "test_invalid_types"
  ],
  "failedSubcases": [
    "True",
    "False"
  ],
  "answerSeconds": 6.166544499999873,
  "coldStartSeconds": 2.58899058300085,
  "peakSampledProcessRssBytes": 5899993088,
  "promptSha256": "b2551654038a7090f4e83287dc47761375a63dc274dbbddcc04298cb38eba37c",
  "acceptanceTestsSha256": "e106c8a64f92e2f84c73cc6a2eeb2d900a0f4365f1e044f1babe8a7a7b20e601",
  "generatedSourceSha256": "900d987d2306f0096338c8f1513fa44ef56ede6f0098e699b1ccc7e7d53120e6",
  "modelSha256": "7485fe6f11af29433bc51cab58009521f205840f5b4ae3a32fa7f92e8534fdf5",
  "binarySha256": "51afbf4ea6485b9ada19e99af97181f12baca42ef704b1e291c26a9ac0644eae",
  "runtime": "version: 0.6.0 (build 11429, commit d81235049)\nbuilt with AppleClang 21.0.0.21000334 for Darwin arm64",
  "hardware": {
    "chip": "Apple M4 Pro",
    "memoryBytes": 25769803776,
    "os": "26.3",
    "build": "25D125"
  },
  "serverFlags": [
    "--alias",
    "kingy-qwen3-4b",
    "--host",
    "127.0.0.1",
    "--port",
    "18482",
    "-c",
    "8192",
    "-np",
    "1",
    "-t",
    "4",
    "-tb",
    "4",
    "-ngl",
    "0",
    "--no-op-offload",
    "--cache-ram",
    "0",
    "--no-cache-prompt",
    "--jinja",
    "--reasoning",
    "off",
    "--reasoning-budget",
    "0",
    "--no-webui",
    "--offline",
    "--no-context-shift",
    "--cache-type-k",
    "f16",
    "--cache-type-v",
    "f16",
    "--temp",
    "0"
  ],
  "requestSettings": {
    "temperature": 0,
    "seed": 42,
    "max_tokens": 512,
    "stream": false,
    "cache_prompt": false,
    "chat_template_kwargs": {
      "enable_thinking": false
    },
    "response_format": {
      "type": "json_schema",
      "json_schema": {
        "name": "billing_patch",
        "strict": true,
        "schema": {
          "type": "object",
          "properties": {
            "billing_py": {
              "type": "string"
            }
          },
          "required": [
            "billing_py"
          ],
          "additionalProperties": false
        }
      }
    }
  },
  "rawResponse": "raw-response.json",
  "failedReplay": "failed-replay.json",
  "originalTestsUnchanged": true,
  "ownedServerStopped": true,
  "ownedKeyRemoved": true,
  "limitations": [
    "One exposed synthetic baseline sample; no unseen/general coding quality claim.",
    "Observed latency and process RSS were measured during a failed task on a shared machine; not corrected-task performance.",
    "No corrected model output has been generated."
  ],
  "sourceProvenance": {
    "originalRunReceiptSha256": "e0f426e727d928b5517b941738a010bb731ed2a3cf8bffd5ad5f4edd7be0fe3b",
    "originalReplaySha256": "129c0d8c2defa548f58c5a60f7c501f838f9f6e07c977effa030e2d278d6c8bc"
  },
  "configuration": {
    "chip": "Apple M4 Pro",
    "memoryGiB": 24,
    "osVersion": "26.3",
    "osBuild": "25D125",
    "runtimeBuild": 11429,
    "runtimeCommit": "d81235049",
    "modelSha256": "7485fe6f11af29433bc51cab58009521f205840f5b4ae3a32fa7f92e8534fdf5",
    "contextTokens": 8192,
    "concurrency": 1,
    "gpuLayers": 0,
    "threads": 4,
    "batchThreads": 4,
    "cacheK": "f16",
    "cacheV": "f16",
    "pythonVersion": "3.14.0"
  },
  "projection": "Public summary of the retained actual .2 failure; temp paths redacted from separate replay projection. Original private receipts remain unchanged. Exact model response is inside the ZIP.",
  "receipt": "coding-baseline-failure-20261030.3.json"
}