{
  "context": {
    "cap_tokens": 64256,
    "max_model_len": 65536,
    "targets": [
      {
        "attempted_context_tokens": 64000,
        "chars_per_token": 3.0,
        "clamped_tokens": 64000,
        "content_chunks": 19,
        "decode_tok_s": 143.59680789826064,
        "e2e_ms": 12924.405799945816,
        "effective_prefill_tok_s": 4988.220475702129,
        "estimated_prompt_tokens": 64000,
        "generation_ms": 403.9086999837309,
        "mean_inter_token_latency_ms": 6.963943103167774,
        "output_token_source": "usage",
        "output_tokens": 59,
        "prompt_tokens": 62455,
        "reasoning_chunks": 0,
        "status": "passed",
        "target_tokens": 64000,
        "time_to_first_output_ms": 12520.497099962085,
        "ttft_ms": 12520.498499972746,
        "usage": {
          "completion_tokens": 59,
          "prompt_tokens": 62455,
          "total_tokens": 62514
        },
        "visible_generation_ms": 403.9072999730706
      }
    ]
  },
  "evaluation_protocol": {
    "budget_semantics": "visible_answer_tokens plus reasoning_headroom_tokens are sent as one max_tokens cap; the endpoint does not hard-partition the channels",
    "minimum_comparable_repetitions": 3,
    "minimum_pass_rate": 1.0,
    "reasoning_headroom_tokens": 0,
    "records_finish_reason": true,
    "records_full_visible_answer": true,
    "records_reasoning_channel_metadata": true,
    "repetitions": 3,
    "version": 3,
    "visible_answer_tokens": 256
  },
  "failures": [],
  "identity": {
    "base_url": "http://127.0.0.1:8080/v1",
    "candidate_id": "qwen38-huihui",
    "config_id": "baked-mtp3-64k",
    "engine": "ninfer",
    "gpu": "RTX5090",
    "model": "qwen38-huihui-ninfer-schemafix-mtp3-64k",
    "started_at": "2026-09-19T23:28:34Z"
  },
  "intelligence": {
    "checks": [
      {
        "attempt_count": 3,
        "attempts": [
          {
            "attempt": 1,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```",
            "content_chars": 94,
            "content_excerpt": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 317.93450005352497,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "text_checks": [
              {
                "name": "diff_shape",
                "passed": true
              },
              {
                "name": "removes_old_timeout",
                "passed": true
              },
              {
                "name": "adds_new_timeout",
                "passed": true
              }
            ],
            "usage": {
              "completion_tokens": 45,
              "prompt_tokens": 51,
              "total_tokens": 96
            }
          },
          {
            "attempt": 2,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```",
            "content_chars": 94,
            "content_excerpt": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 293.4961999999359,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "text_checks": [
              {
                "name": "diff_shape",
                "passed": true
              },
              {
                "name": "removes_old_timeout",
                "passed": true
              },
              {
                "name": "adds_new_timeout",
                "passed": true
              }
            ],
            "usage": {
              "completion_tokens": 45,
              "prompt_tokens": 51,
              "total_tokens": 96
            }
          },
          {
            "attempt": 3,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```",
            "content_chars": 94,
            "content_excerpt": "```diff\n--- a/app.py\n+++ b/app.py\n@@ -1,2 +1,2 @@\n-timeout = 30\n+timeout = 45\n retries = 2\n```",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 293.8708000583574,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "text_checks": [
              {
                "name": "diff_shape",
                "passed": true
              },
              {
                "name": "removes_old_timeout",
                "passed": true
              },
              {
                "name": "adds_new_timeout",
                "passed": true
              }
            ],
            "usage": {
              "completion_tokens": 45,
              "prompt_tokens": 51,
              "total_tokens": 96
            }
          }
        ],
        "id": "unified_diff_timeout_edit",
        "pass_count": 3,
        "pass_rate": 1.0,
        "required_pass_rate": 1.0,
        "status": "passed",
        "validator": "deterministic_text_checks"
      },
      {
        "attempt_count": 3,
        "attempts": [
          {
            "attempt": 1,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "The total latency of 3000 ms exceeds the 2500 ms timeout because the LLM takes too long, so you should reduce the LLM's response time by using a faster model or limiting its output length.",
            "content_chars": 188,
            "content_excerpt": "The total latency of 3000 ms exceeds the 2500 ms timeout because the LLM takes too long, so you should reduce the LLM's response time by using a faster model or limiting its output length.",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 442.8418999304995,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "text_checks": [
              {
                "name": "identifies_budget_overrun",
                "passed": true
              },
              {
                "name": "offers_latency_fix",
                "passed": true
              }
            ],
            "usage": {
              "completion_tokens": 49,
              "prompt_tokens": 79,
              "total_tokens": 128
            }
          },
          {
            "attempt": 2,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "The total latency of 3000 ms exceeds the 2500 ms timeout because the LLM takes too long, so you should reduce the LLM's response time by using a faster model or limiting its output length.",
            "content_chars": 188,
            "content_excerpt": "The total latency of 3000 ms exceeds the 2500 ms timeout because the LLM takes too long, so you should reduce the LLM's response time by using a faster model or limiting its output length.",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 410.55040003266186,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "text_checks": [
              {
                "name": "identifies_budget_overrun",
                "passed": true
              },
              {
                "name": "offers_latency_fix",
                "passed": true
              }
            ],
            "usage": {
              "completion_tokens": 49,
              "prompt_tokens": 79,
              "total_tokens": 128
            }
          },
          {
            "attempt": 3,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "The total latency of 3000 ms exceeds the 2500 ms timeout because the LLM takes too long, so you should reduce the LLM's response time by using a faster model or limiting its output length.",
            "content_chars": 188,
            "content_excerpt": "The total latency of 3000 ms exceeds the 2500 ms timeout because the LLM takes too long, so you should reduce the LLM's response time by using a faster model or limiting its output length.",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 410.32839997205883,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "text_checks": [
              {
                "name": "identifies_budget_overrun",
                "passed": true
              },
              {
                "name": "offers_latency_fix",
                "passed": true
              }
            ],
            "usage": {
              "completion_tokens": 49,
              "prompt_tokens": 79,
              "total_tokens": 128
            }
          }
        ],
        "id": "parallel_timeout_triage",
        "pass_count": 3,
        "pass_rate": 1.0,
        "required_pass_rate": 1.0,
        "status": "passed",
        "validator": "deterministic_text_checks"
      }
    ],
    "status": "passed"
  },
  "run_id": "fast-bakeoff-20260919T232834Z",
  "sampling": {
    "temperature": {
      "effective_request": 0.0,
      "requested": null,
      "sent": true
    },
    "top_p": {
      "effective_request": null,
      "requested": null,
      "sent": false
    }
  },
  "schema": "anvil-serving.fast-tier-bakeoff/v1",
  "score_inputs": {
    "e2e_p50_ms": 12924.405799945816,
    "intelligence_pass_rate": 1.0,
    "operational_fit_notes": [
      "endpoint was already loaded; benchmark did not start or stop serves"
    ],
    "session_recall_passed": true,
    "thinking_mode": "unsupported",
    "tool_call_passed": true,
    "ttft_p50_ms": 12520.498499972746,
    "usable_context_tokens": 64000,
    "voice_latency_ms": null
  },
  "selection": {
    "context_targets": [
      64000
    ],
    "endpoint_already_loaded": true,
    "requests_per_context": 1,
    "suites": [
      "tool",
      "intelligence",
      "session",
      "context"
    ]
  },
  "session": {
    "checks": [
      {
        "attempt_count": 3,
        "attempts": [
          {
            "attempt": 1,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "RIVER-918",
            "content_chars": 9,
            "content_excerpt": "RIVER-918",
            "expected": "RIVER-918",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 143.41099990997463,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "usage": {
              "completion_tokens": 7,
              "prompt_tokens": 52,
              "total_tokens": 59
            }
          },
          {
            "attempt": 2,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "RIVER-918",
            "content_chars": 9,
            "content_excerpt": "RIVER-918",
            "expected": "RIVER-918",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 79.1169999865815,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "usage": {
              "completion_tokens": 7,
              "prompt_tokens": 52,
              "total_tokens": 59
            }
          },
          {
            "attempt": 3,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "RIVER-918",
            "content_chars": 9,
            "content_excerpt": "RIVER-918",
            "expected": "RIVER-918",
            "failure_class": null,
            "finish_reason": "stop",
            "latency_ms": 78.69350002147257,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "usage": {
              "completion_tokens": 7,
              "prompt_tokens": 52,
              "total_tokens": 59
            }
          }
        ],
        "name": "single_request_multiturn_recall",
        "pass_count": 3,
        "pass_rate": 1.0,
        "required_pass_rate": 1.0,
        "status": "passed"
      }
    ],
    "status": "passed"
  },
  "source_recipe": {
    "ref": null,
    "serve_command": null
  },
  "suites": {},
  "thinking": {
    "chat_template_kwargs": null,
    "control_evidence": null,
    "control_evidence_sha256": null,
    "control_mechanism": "unsupported",
    "control_requested": null,
    "control_status": "unsupported",
    "mode": "unsupported",
    "reasoning_effort": null,
    "unsupported": true
  },
  "timing": {
    "chat": {
      "completion_tokens_max": 59.0,
      "completion_tokens_mean": 59.0,
      "completion_tokens_mean_ci95_high": 59.0,
      "completion_tokens_mean_ci95_low": 59.0,
      "completion_tokens_min": 59.0,
      "completion_tokens_p25": 59.0,
      "completion_tokens_p50": 59.0,
      "completion_tokens_p75": 59.0,
      "completion_tokens_p90": 59.0,
      "completion_tokens_p95": 59.0,
      "completion_tokens_p99": 59.0,
      "completion_tokens_samples": 1,
      "completion_tokens_stddev": 0.0,
      "content_chunks": 19,
      "controlled_output_exact_adherence": 0,
      "controlled_output_exact_adherence_rate": null,
      "controlled_output_samples": 0,
      "controlled_output_unobservable": 0,
      "decode_tok_s_max": 143.59680789826064,
      "decode_tok_s_mean": 143.59680789826064,
      "decode_tok_s_mean_ci95_high": 143.59680789826064,
      "decode_tok_s_mean_ci95_low": 143.59680789826064,
      "decode_tok_s_min": 143.59680789826064,
      "decode_tok_s_p25": 143.59680789826064,
      "decode_tok_s_p50": 143.59680789826064,
      "decode_tok_s_p75": 143.59680789826064,
      "decode_tok_s_p90": 143.59680789826064,
      "decode_tok_s_p95": 143.59680789826064,
      "decode_tok_s_p99": 143.59680789826064,
      "decode_tok_s_samples": 1,
      "decode_tok_s_stddev": 0.0,
      "e2e_max_ms": 12924.405799945816,
      "e2e_mean_ci95_high_ms": 12924.405799945816,
      "e2e_mean_ci95_low_ms": 12924.405799945816,
      "e2e_mean_ms": 12924.405799945816,
      "e2e_min_ms": 12924.405799945816,
      "e2e_p25_ms": 12924.405799945816,
      "e2e_p50_ms": 12924.405799945816,
      "e2e_p75_ms": 12924.405799945816,
      "e2e_p90_ms": 12924.405799945816,
      "e2e_p95_ms": 12924.405799945816,
      "e2e_p99_ms": 12924.405799945816,
      "e2e_samples": 1,
      "e2e_stddev_ms": 0.0,
      "effective_prefill_tok_s_max": 4988.220475702129,
      "effective_prefill_tok_s_mean": 4988.220475702129,
      "effective_prefill_tok_s_mean_ci95_high": 4988.220475702129,
      "effective_prefill_tok_s_mean_ci95_low": 4988.220475702129,
      "effective_prefill_tok_s_min": 4988.220475702129,
      "effective_prefill_tok_s_p25": 4988.220475702129,
      "effective_prefill_tok_s_p50": 4988.220475702129,
      "effective_prefill_tok_s_p75": 4988.220475702129,
      "effective_prefill_tok_s_p90": 4988.220475702129,
      "effective_prefill_tok_s_p95": 4988.220475702129,
      "effective_prefill_tok_s_p99": 4988.220475702129,
      "effective_prefill_tok_s_samples": 1,
      "effective_prefill_tok_s_stddev": 0.0,
      "generation_max_ms": 403.9086999837309,
      "generation_mean_ci95_high_ms": 403.9086999837309,
      "generation_mean_ci95_low_ms": 403.9086999837309,
      "generation_mean_ms": 403.9086999837309,
      "generation_min_ms": 403.9086999837309,
      "generation_p25_ms": 403.9086999837309,
      "generation_p50_ms": 403.9086999837309,
      "generation_p75_ms": 403.9086999837309,
      "generation_p90_ms": 403.9086999837309,
      "generation_p95_ms": 403.9086999837309,
      "generation_p99_ms": 403.9086999837309,
      "generation_samples": 1,
      "generation_stddev_ms": 0.0,
      "mean_inter_token_latency_ms_max": 6.963943103167774,
      "mean_inter_token_latency_ms_mean": 6.963943103167774,
      "mean_inter_token_latency_ms_mean_ci95_high": 6.963943103167774,
      "mean_inter_token_latency_ms_mean_ci95_low": 6.963943103167774,
      "mean_inter_token_latency_ms_min": 6.963943103167774,
      "mean_inter_token_latency_ms_p25": 6.963943103167774,
      "mean_inter_token_latency_ms_p50": 6.963943103167774,
      "mean_inter_token_latency_ms_p75": 6.963943103167774,
      "mean_inter_token_latency_ms_p90": 6.963943103167774,
      "mean_inter_token_latency_ms_p95": 6.963943103167774,
      "mean_inter_token_latency_ms_p99": 6.963943103167774,
      "mean_inter_token_latency_ms_samples": 1,
      "mean_inter_token_latency_ms_stddev": 0.0,
      "mean_time_per_output_token_max_ms": 6.963943103167774,
      "mean_time_per_output_token_mean_ci95_high_ms": 6.963943103167774,
      "mean_time_per_output_token_mean_ci95_low_ms": 6.963943103167774,
      "mean_time_per_output_token_mean_ms": 6.963943103167774,
      "mean_time_per_output_token_min_ms": 6.963943103167774,
      "mean_time_per_output_token_p25_ms": 6.963943103167774,
      "mean_time_per_output_token_p50_ms": 6.963943103167774,
      "mean_time_per_output_token_p75_ms": 6.963943103167774,
      "mean_time_per_output_token_p90_ms": 6.963943103167774,
      "mean_time_per_output_token_p95_ms": 6.963943103167774,
      "mean_time_per_output_token_p99_ms": 6.963943103167774,
      "mean_time_per_output_token_samples": 1,
      "mean_time_per_output_token_stddev_ms": 0.0,
      "output_tokens": 59,
      "prompt_token_samples": 1,
      "prompt_tokens": 62455,
      "prompt_tokens_max": 62455.0,
      "prompt_tokens_mean": 62455.0,
      "prompt_tokens_mean_ci95_high": 62455.0,
      "prompt_tokens_mean_ci95_low": 62455.0,
      "prompt_tokens_min": 62455.0,
      "prompt_tokens_p25": 62455.0,
      "prompt_tokens_p50": 62455.0,
      "prompt_tokens_p75": 62455.0,
      "prompt_tokens_p90": 62455.0,
      "prompt_tokens_p95": 62455.0,
      "prompt_tokens_p99": 62455.0,
      "prompt_tokens_samples": 1,
      "prompt_tokens_stddev": 0.0,
      "reasoning_chunks": 0,
      "request_canary_passed": 0,
      "request_canary_samples": 0,
      "time_to_first_output_max_ms": 12520.497099962085,
      "time_to_first_output_mean_ci95_high_ms": 12520.497099962085,
      "time_to_first_output_mean_ci95_low_ms": 12520.497099962085,
      "time_to_first_output_mean_ms": 12520.497099962085,
      "time_to_first_output_min_ms": 12520.497099962085,
      "time_to_first_output_p25_ms": 12520.497099962085,
      "time_to_first_output_p50_ms": 12520.497099962085,
      "time_to_first_output_p75_ms": 12520.497099962085,
      "time_to_first_output_p90_ms": 12520.497099962085,
      "time_to_first_output_p95_ms": 12520.497099962085,
      "time_to_first_output_p99_ms": 12520.497099962085,
      "time_to_first_output_samples": 1,
      "time_to_first_output_stddev_ms": 0.0,
      "tpot_max_ms": 6.963943103167774,
      "tpot_mean_ci95_high_ms": 6.963943103167774,
      "tpot_mean_ci95_low_ms": 6.963943103167774,
      "tpot_mean_ms": 6.963943103167774,
      "tpot_min_ms": 6.963943103167774,
      "tpot_p25_ms": 6.963943103167774,
      "tpot_p50_ms": 6.963943103167774,
      "tpot_p75_ms": 6.963943103167774,
      "tpot_p90_ms": 6.963943103167774,
      "tpot_p95_ms": 6.963943103167774,
      "tpot_p99_ms": 6.963943103167774,
      "tpot_samples": 1,
      "tpot_stddev_ms": 0.0,
      "ttft_max_ms": 12520.498499972746,
      "ttft_mean_ci95_high_ms": 12520.498499972746,
      "ttft_mean_ci95_low_ms": 12520.498499972746,
      "ttft_mean_ms": 12520.498499972746,
      "ttft_min_ms": 12520.498499972746,
      "ttft_p25_ms": 12520.498499972746,
      "ttft_p50_ms": 12520.498499972746,
      "ttft_p75_ms": 12520.498499972746,
      "ttft_p90_ms": 12520.498499972746,
      "ttft_p95_ms": 12520.498499972746,
      "ttft_p99_ms": 12520.498499972746,
      "ttft_samples": 1,
      "ttft_stddev_ms": 0.0
    },
    "wall_ms": 16065.857500070706
  },
  "tool": {
    "checks": [
      {
        "attempt_count": 3,
        "attempts": [
          {
            "arguments": {
              "zip": "98101"
            },
            "attempt": 1,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "",
            "content_chars": 0,
            "content_excerpt": "",
            "failure_class": null,
            "finish_reason": "tool_calls",
            "latency_ms": 274.9343999894336,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "tool_argument_observations": [
              {
                "arguments_chars": 15,
                "arguments_excerpt": "{\"zip\":\"98101\"}",
                "arguments_truncated": false,
                "arguments_wire_type": "str",
                "name": "record_weather_zip"
              }
            ],
            "tool_call_count": 1,
            "usage": {
              "completion_tokens": 32,
              "prompt_tokens": 292,
              "total_tokens": 324
            },
            "valid_tool_call_count": 1,
            "validation_errors": []
          },
          {
            "arguments": {
              "zip": "98101"
            },
            "attempt": 2,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "",
            "content_chars": 0,
            "content_excerpt": "",
            "failure_class": null,
            "finish_reason": "tool_calls",
            "latency_ms": 196.92719995509833,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "tool_argument_observations": [
              {
                "arguments_chars": 15,
                "arguments_excerpt": "{\"zip\":\"98101\"}",
                "arguments_truncated": false,
                "arguments_wire_type": "str",
                "name": "record_weather_zip"
              }
            ],
            "tool_call_count": 1,
            "usage": {
              "completion_tokens": 32,
              "prompt_tokens": 292,
              "total_tokens": 324
            },
            "valid_tool_call_count": 1,
            "validation_errors": []
          },
          {
            "arguments": {
              "zip": "98101"
            },
            "attempt": 3,
            "budget": {
              "legacy_max_tokens_as_visible": false,
              "max_completion_tokens": 256,
              "reasoning_headroom_tokens": 0,
              "visible_answer_tokens": 256
            },
            "content": "",
            "content_chars": 0,
            "content_excerpt": "",
            "failure_class": null,
            "finish_reason": "tool_calls",
            "latency_ms": 196.900499984622,
            "reasoning_chars": 0,
            "reasoning_excerpt": "",
            "reasoning_field": null,
            "reasoning_tokens": null,
            "status": "passed",
            "tool_argument_observations": [
              {
                "arguments_chars": 15,
                "arguments_excerpt": "{\"zip\":\"98101\"}",
                "arguments_truncated": false,
                "arguments_wire_type": "str",
                "name": "record_weather_zip"
              }
            ],
            "tool_call_count": 1,
            "usage": {
              "completion_tokens": 32,
              "prompt_tokens": 292,
              "total_tokens": 324
            },
            "valid_tool_call_count": 1,
            "validation_errors": []
          }
        ],
        "expected_arguments": {
          "zip": "98101"
        },
        "expected_function": "record_weather_zip",
        "name": "openai_tool_call_smoke",
        "pass_count": 3,
        "pass_rate": 1.0,
        "required_pass_rate": 1.0,
        "status": "passed"
      }
    ],
    "status": "passed"
  },
  "voice": {
    "status": "not_run",
    "stt_latency_ms": null,
    "total_turn_latency_ms": null,
    "tts_latency_ms": null
  }
}
