{
  "created_at": "2026-10-26T13:45:28.394397Z",
  "failure": null,
  "partial": false,
  "provenance": {
    "endpoint": {
      "auth_env": "ANVIL_EVALUATION_TOKEN",
      "base_url": "http://127.0.0.1:8000/v1",
      "model": "llm.evaluation.ignis"
    },
    "finished_at": "2026-10-26T13:45:28.394301Z",
    "started_at": "2026-10-26T13:43:11.727340Z",
    "submitted_at": "2026-10-26T13:34:06.578529Z",
    "worker": {
      "id": "Primary Node"
    }
  },
  "results": {
    "evidence": {
      "completeness": "completed",
      "created_at": "2026-10-26T13:45:28.380731Z",
      "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": "sha256:0226b62047113c219124ca4a02b1e52492402b44f5a3020db91e77756914b8cc"
        },
        "model": {
          "repository": "gpillon/Qwen3.8-Flash-Next-trellis-a25-ignis",
          "revision": "8d94db83cd8e58d6be31e796723e5e35b960638c"
        },
        "runtime": {
          "engine": "ignis-0.8.0",
          "recipe_sha256": "b5135421c217552c18bfffdcc2bbbd0e63e6d9e93f167d2014dc5a56c0fe47b0",
          "source_revision": "950ee0d1c2cf6178cd7fec29df3f17c3e127f987"
        },
        "served_model": {
          "configured_context_tokens": 131072,
          "id": "qwen3.8-flash-next",
          "router_alias": "llm.evaluation.ignis"
        },
        "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": "ignis080-context-scout-r1-20261026",
        "spec_sha256": "4fe6e8d156b331bdf8c8474e747757ce3e62bce153817559059728ca8d89eec4",
        "suite": "context"
      },
      "schema": "anvil-serving.benchmark-evidence/v1",
      "stages": [
        {
          "evidence": [
            {
              "bytes": 586,
              "path": "evidence/0-assets.json",
              "sha256": "947cd37a10dfbec99c76bd40571516adf448877832def500feeb40f10fd7b491"
            }
          ],
          "name": "asset_preparation",
          "sequence": 0,
          "status": "completed"
        },
        {
          "evidence": [
            {
              "bytes": 2211,
              "path": "evidence/0-preflight.json",
              "sha256": "b7d27070be47e67f9470ff6d972efbc66c70a916e06e8e742085d782099a7755"
            }
          ],
          "name": "preflight",
          "sequence": 1,
          "status": "completed"
        },
        {
          "evidence": [
            {
              "bytes": 28816,
              "path": "evidence/0-context.json",
              "sha256": "f113aa0ef5490c3dea02560f2cc33b6aa9ac094ce430568db3b8b4c45174592e"
            }
          ],
          "name": "context",
          "sequence": 2,
          "status": "completed"
        }
      ],
      "summary": {
        "advertised_context": 131072,
        "attempted_buckets": [
          8192,
          32768,
          114688
        ],
        "buckets": [
          {
            "completed_count": 3,
            "completion_rate": 1.0,
            "engine_telemetry": {
              "available": false,
              "observations": []
            },
            "failures": {},
            "latency_ms": {
              "available": true,
              "mean": 2861.747172040244,
              "observations": [
                2620.845308061689,
                3063.134313095361,
                2901.2618949636817
              ]
            },
            "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": 2620.845308061689,
                "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": 40.82652252342756,
                "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": 3063.134313095361,
                "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": 62.680894918375294,
                "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": 2901.2618949636817,
                "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": 51.35696307136216,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              }
            ],
            "throughput_tps": {
              "available": true,
              "mean": 51.62146017105501,
              "observations": [
                40.82652252342756,
                62.680894918375294,
                51.35696307136216
              ]
            }
          },
          {
            "completed_count": 3,
            "completion_rate": 1.0,
            "engine_telemetry": {
              "available": false,
              "observations": []
            },
            "failures": {},
            "latency_ms": {
              "available": true,
              "mean": 9267.942331653709,
              "observations": [
                9011.548401089385,
                9403.395975008607,
                9388.882618863136
              ]
            },
            "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": 9011.548401089385,
                "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": 13.42721523699219,
                "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": 9403.395975008607,
                "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": 21.481600959574088,
                "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": 9388.882618863136,
                "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": 17.786994126913623,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              }
            ],
            "throughput_tps": {
              "available": true,
              "mean": 17.565270107826635,
              "observations": [
                13.42721523699219,
                21.481600959574088,
                17.786994126913623
              ]
            }
          },
          {
            "completed_count": 3,
            "completion_rate": 1.0,
            "engine_telemetry": {
              "available": false,
              "observations": []
            },
            "failures": {},
            "latency_ms": {
              "available": true,
              "mean": 32497.507145317893,
              "observations": [
                32019.511128077284,
                32671.336503932253,
                32801.67380394414
              ]
            },
            "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": 32019.511128077284,
                "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": 4.122486426229406,
                "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": 32671.336503932253,
                "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": 6.0603581361346865,
                "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": 32801.67380394414,
                "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": 4.816827365102379,
                "token_measurement": "usage",
                "visible_answer": "[REDACTED_PUBLIC_RESPONSE_TEXT]",
                "visible_answer_truncated": false
              }
            ],
            "throughput_tps": {
              "available": true,
              "mean": 4.9998906424888245,
              "observations": [
                4.122486426229406,
                6.0603581361346865,
                4.816827365102379
              ]
            }
          }
        ],
        "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": "ignis080-context-scout-r1-20261026",
    "spec_sha256": "4fe6e8d156b331bdf8c8474e747757ce3e62bce153817559059728ca8d89eec4",
    "suite": "context"
  },
  "schema": "anvil-serving.benchmark-result/v1",
  "status": "completed"
}
