{
  "created_at": "2026-10-26T15:29:31.627920Z",
  "failure": null,
  "partial": false,
  "provenance": {
    "endpoint": {
      "auth_env": "ANVIL_EVALUATION_TOKEN",
      "base_url": "http://127.0.0.1:8000/v1",
      "model": "llm.evaluation.vllm"
    },
    "finished_at": "2026-10-26T15:29:31.627816Z",
    "started_at": "2026-10-26T15:28:38.160985Z",
    "submitted_at": "2026-10-26T15:28:37.954859+00:00",
    "worker": {
      "id": "Primary Node"
    }
  },
  "results": {
    "evidence": {
      "completeness": "completed",
      "created_at": "2026-10-26T15:29:31.613732Z",
      "evidence_kind": "measured",
      "failure": null,
      "identities": {
        "concurrency": {
          "requests": 1,
          "workers": 1
        },
        "context": {
          "configured_tokens": null,
          "output_headroom_tokens": 8192,
          "profile_buckets": [
            8192,
            32768,
            131072,
            262144
          ],
          "selected_buckets": [
            8192,
            32768,
            114688
          ]
        },
        "dataset": {
          "name": "deterministic-native",
          "profile_sha256": "fce0e6e1e7497585bdfc6f76507d3baaff99a6b4f174d841d268de20c68de7f6"
        },
        "hardware": {
          "gpu_name": "NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition",
          "gpu_uuids": [
            "[REDACTED_PUBLIC_HARDWARE_ID]"
          ]
        },
        "harnesses": {
          "native": {
            "schema": "anvil-serving.context/v1"
          }
        },
        "image": {
          "id": "sha256:bc48d33ee167fb7675a59889fe0a7fea539bc6700c53cada89284d594a754b54"
        },
        "model": {
          "repository": "local-inference-lab/Qwen3.8-Flash-Next-NVFP4-MXFP8-CSF-QAD",
          "revision": "dc5b8ab2fa14bc4fecd6886273b7ebc6095662e2"
        },
        "runtime": {
          "engine": "vllm-karmic-kraken-beta-20261020",
          "engine_full_source_revision": "unknown",
          "engine_version_observed": "v0.1.dev21836+g6d66eed35",
          "publisher_source_revision": "8bb0967ca1170f6e77c5bfdefcbeb85ec0c9a868",
          "recipe_sha256": "58c64c609c50647c3a00b08baf1256f076a0b90cab6bbdf7cb98e4efc8254dff"
        },
        "served_model": {
          "configured_context_tokens": 262144,
          "id": "qwen38-flash-next-vllm-qad-tp1-c16-262k-mtp3",
          "router_alias": "llm.evaluation.vllm"
        },
        "topology": {
          "kind": "co-resident-benchmark-client",
          "worker_id": "Primary Node"
        }
      },
      "promotion": {
        "authorized": false,
        "message": "Benchmark evidence does not authorize model promotion; promotion is a separate human decision."
      },
      "run": {
        "ownership_id": "evaluation-platform",
        "profile": "scout",
        "run_id": "vllmqad-r2-context-scout-r1-20261026",
        "spec_sha256": "f05f5ca7e5bbee2f3137ce5976d75fd90a1d70b2f6af2043dda31746050f252d",
        "suite": "context"
      },
      "schema": "anvil-serving.benchmark-evidence/v1",
      "stages": [
        {
          "evidence": [
            {
              "bytes": 588,
              "path": "evidence/0-assets.json",
              "sha256": "8ccfff07a4da795528efff0863988c7692f2bf53bd14dbff71c65a004472a2aa"
            }
          ],
          "name": "asset_preparation",
          "sequence": 0,
          "status": "completed"
        },
        {
          "evidence": [
            {
              "bytes": 2212,
              "path": "evidence/0-preflight.json",
              "sha256": "171d08e4f2b90f0906149b16bf4916e20cde8a31bdb50bef5d37da0ecb7b47fe"
            }
          ],
          "name": "preflight",
          "sequence": 1,
          "status": "completed"
        },
        {
          "evidence": [
            {
              "bytes": 28000,
              "path": "evidence/0-context.json",
              "sha256": "fcbf2f51ab3e805454a21ca886c404ba166fc2ab3a06709bc2d6acc8293fad18"
            }
          ],
          "name": "context",
          "sequence": 2,
          "status": "completed"
        }
      ],
      "summary": {
        "advertised_context": 262144,
        "attempted_buckets": [
          8192,
          32768,
          114688
        ],
        "buckets": [
          {
            "completed_count": 3,
            "completion_rate": 1.0,
            "engine_telemetry": {
              "available": false,
              "observations": []
            },
            "failures": {},
            "latency_ms": {
              "available": true,
              "mean": 1371.3083560578525,
              "observations": [
                1275.306556141004,
                1470.3913789708167,
                1368.2271330617368
              ]
            },
            "meets_policy": true,
            "pass_rate": 1.0,
            "passed_count": 3,
            "relative_drop_from_baseline": 0.0,
            "requested_tokens": 8192,
            "sample_count": 3,
            "samples": [
              {
                "answer_sha256": "165c06e1078f0d23d51237f55fdf201897c11ab1ba6b89b4be0058e8551485db",
                "case_id": "native-needle-8192-0.500-r0",
                "case_type": "native-needle",
                "completed": true,
                "distractor_count": 0,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 1275.306556141004,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 8198,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 8192,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 1,
                "throughput_tps": 90.95852243636901,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              },
              {
                "answer_sha256": "ea77d2c40d5ae22b614345bf65faaf4a798b6926e2753eaa34d64261ffa8f4cb",
                "case_id": "native-order-8192-0.500-r0",
                "case_type": "native-order",
                "completed": true,
                "distractor_count": 0,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 1470.3913789708167,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 8202,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 8192,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 3,
                "throughput_tps": 129.8973883631501,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              },
              {
                "answer_sha256": "165c06e1078f0d23d51237f55fdf201897c11ab1ba6b89b4be0058e8551485db",
                "case_id": "native-distractor-8192-0.500-r0",
                "case_type": "native-distractor",
                "completed": true,
                "distractor_count": 2,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 1368.2271330617368,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 8208,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 8192,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 1,
                "throughput_tps": 115.4779028876858,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              }
            ],
            "throughput_tps": {
              "available": true,
              "mean": 112.1112712290683,
              "observations": [
                90.95852243636901,
                129.8973883631501,
                115.4779028876858
              ]
            }
          },
          {
            "completed_count": 3,
            "completion_rate": 1.0,
            "engine_telemetry": {
              "available": false,
              "observations": []
            },
            "failures": {},
            "latency_ms": {
              "available": true,
              "mean": 3459.9548026453704,
              "observations": [
                3303.5518750548363,
                3592.6947679836303,
                3483.6177648976445
              ]
            },
            "meets_policy": true,
            "pass_rate": 1.0,
            "passed_count": 3,
            "relative_drop_from_baseline": 0.0,
            "requested_tokens": 32768,
            "sample_count": 3,
            "samples": [
              {
                "answer_sha256": "165c06e1078f0d23d51237f55fdf201897c11ab1ba6b89b4be0058e8551485db",
                "case_id": "native-needle-32768-0.500-r0",
                "case_type": "native-needle",
                "completed": true,
                "distractor_count": 0,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 3303.5518750548363,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 32658,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 32768,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 1,
                "throughput_tps": 29.6650404493416,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              },
              {
                "answer_sha256": "ea77d2c40d5ae22b614345bf65faaf4a798b6926e2753eaa34d64261ffa8f4cb",
                "case_id": "native-order-32768-0.500-r0",
                "case_type": "native-order",
                "completed": true,
                "distractor_count": 0,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 3592.6947679836303,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 32662,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 32768,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 3,
                "throughput_tps": 53.441779054266384,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              },
              {
                "answer_sha256": "165c06e1078f0d23d51237f55fdf201897c11ab1ba6b89b4be0058e8551485db",
                "case_id": "native-distractor-32768-0.500-r0",
                "case_type": "native-distractor",
                "completed": true,
                "distractor_count": 2,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 3483.6177648976445,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 32668,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 32768,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 1,
                "throughput_tps": 45.35514820026253,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              }
            ],
            "throughput_tps": {
              "available": true,
              "mean": 42.82065590129017,
              "observations": [
                29.6650404493416,
                53.441779054266384,
                45.35514820026253
              ]
            }
          },
          {
            "completed_count": 3,
            "completion_rate": 1.0,
            "engine_telemetry": {
              "available": false,
              "observations": []
            },
            "failures": {},
            "latency_ms": {
              "available": true,
              "mean": 12636.447718599811,
              "observations": [
                12301.28965084441,
                12830.071042990312,
                12777.982461964712
              ]
            },
            "meets_policy": true,
            "pass_rate": 1.0,
            "passed_count": 3,
            "relative_drop_from_baseline": 0.0,
            "requested_tokens": 114688,
            "sample_count": 3,
            "samples": [
              {
                "answer_sha256": "165c06e1078f0d23d51237f55fdf201897c11ab1ba6b89b4be0058e8551485db",
                "case_id": "native-needle-114688-0.500-r0",
                "case_type": "native-needle",
                "completed": true,
                "distractor_count": 0,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 12301.28965084441,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 114578,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 114688,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 1,
                "throughput_tps": 6.340798584044866,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              },
              {
                "answer_sha256": "ea77d2c40d5ae22b614345bf65faaf4a798b6926e2753eaa34d64261ffa8f4cb",
                "case_id": "native-order-114688-0.500-r0",
                "case_type": "native-order",
                "completed": true,
                "distractor_count": 0,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 12830.071042990312,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 114582,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 114688,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 3,
                "throughput_tps": 15.27661065501927,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              },
              {
                "answer_sha256": "165c06e1078f0d23d51237f55fdf201897c11ab1ba6b89b4be0058e8551485db",
                "case_id": "native-distractor-114688-0.500-r0",
                "case_type": "native-distractor",
                "completed": true,
                "distractor_count": 2,
                "engine_telemetry": null,
                "failure": null,
                "finish_reason": "stop",
                "latency_ms": 12777.982461964712,
                "passed": true,
                "position": 0.5,
                "prompt_tokens": 114588,
                "raw_reasoning": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "raw_reasoning_field": "reasoning_content",
                "raw_reasoning_truncated": false,
                "requested_tokens": 114688,
                "schema": "anvil-serving.context-observation/v1",
                "target_count": 1,
                "throughput_tps": 12.443278934940137,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              }
            ],
            "throughput_tps": {
              "available": true,
              "mean": 11.353562724668093,
              "observations": [
                6.340798584044866,
                15.27661065501927,
                12.443278934940137
              ]
            }
          }
        ],
        "effective_context": 114688,
        "first_material_degradation": null,
        "notes": [
          "effective_context uses attempted buckets only",
          "missing engine telemetry is unavailable and is not inferred from latency"
        ],
        "schema": "anvil-serving.context-curve/v1",
        "threshold_policy": {
          "baseline_bucket": 8192,
          "baseline_pass_rate": 1.0,
          "max_relative_drop": 0.25,
          "pass_rate_floor": 0.75
        }
      }
    }
  },
  "run": {
    "ownership_id": "evaluation-platform",
    "profile": "scout",
    "run_id": "vllmqad-r2-context-scout-r1-20261026",
    "spec_sha256": "f05f5ca7e5bbee2f3137ce5976d75fd90a1d70b2f6af2043dda31746050f252d",
    "suite": "context"
  },
  "schema": "anvil-serving.benchmark-result/v1",
  "status": "completed"
}
