{
  "audited_utc": "2026-09-24T20:04:14.628020+00:00",
  "complete": true,
  "planned": 198,
  "observed": 198,
  "known_model_usd": 4.150898000000002,
  "remaining_episodes": 0,
  "budget_usd": 5.0,
  "generation_requests": 338,
  "scored_model_requests": 336,
  "warmup_requests": 2,
  "preflight_not_sent": 0,
  "unknown_usage_requests": [],
  "input_tokens": 1392,
  "cache_read_input_tokens": 548125,
  "cache_creation_input_tokens": 517437,
  "shared_cache_setup_usd": 0.006072,
  "cache_counts_by_arm": {
    "omnilink_parser": {
      "input_tokens": 160,
      "cache_read_input_tokens": 52741,
      "cache_creation_input_tokens": 61922,
      "output_tokens": 8881
    },
    "plain": {
      "input_tokens": 284,
      "cache_read_input_tokens": 121075,
      "cache_creation_input_tokens": 109085,
      "output_tokens": 15424
    },
    "langgraph": {
      "input_tokens": 288,
      "cache_read_input_tokens": 132862,
      "cache_creation_input_tokens": 110940,
      "output_tokens": 15113
    },
    "lobster": {
      "input_tokens": 288,
      "cache_read_input_tokens": 133016,
      "cache_creation_input_tokens": 112361,
      "output_tokens": 15587
    },
    "langgraph_parser": {
      "input_tokens": 164,
      "cache_read_input_tokens": 55726,
      "cache_creation_input_tokens": 61523,
      "output_tokens": 8665
    },
    "lobster_parser": {
      "input_tokens": 160,
      "cache_read_input_tokens": 52705,
      "cache_creation_input_tokens": 60430,
      "output_tokens": 8756
    }
  },
  "uncached_list_price_equivalent_for_recorded_scored_tokens_usd": 5.71144,
  "cache_cost_saving_against_same_token_equivalent_usd": 1.5605419999999972,
  "usd_per_success_with_equal_setup_allocation": {
    "omnilink_parser": 0.015134248484848484,
    "plain": 0.027508375,
    "langgraph": 0.02683928484848485,
    "lobster": 0.02734279393939394,
    "langgraph_parser": 0.014961460606060606,
    "lobster_parser": 0.014832212121212121
  },
  "output_tokens": 72426,
  "truncated_responses": 0,
  "returned_models": [
    "claude-opus-5-5"
  ],
  "summary": {
    "omnilink_parser": {
      "episodes": 33,
      "pass": 33,
      "fail": 0,
      "error": 0,
      "success_rate": 1.0,
      "unsafe": 0,
      "unknown_safety": 0,
      "derived_usd": 0.4984182,
      "usd_per_success": 0.015103581818181817,
      "elapsed_s": 204.25000000000728,
      "seconds_per_success": 6.18939393939416,
      "model_calls": 40
    },
    "plain": {
      "episodes": 33,
      "pass": 32,
      "fail": 1,
      "error": 0,
      "success_rate": 0.9696969696969697,
      "unsafe": 0,
      "unknown_safety": 0,
      "derived_usd": 0.879256,
      "usd_per_success": 0.02747675,
      "elapsed_s": 340.720000000023,
      "seconds_per_success": 10.647500000000719,
      "model_calls": 71
    },
    "langgraph": {
      "episodes": 33,
      "pass": 33,
      "fail": 0,
      "error": 0,
      "success_rate": 1.0,
      "unsafe": 0,
      "unknown_safety": 0,
      "derived_usd": 0.8846844,
      "usd_per_success": 0.02680861818181818,
      "elapsed_s": 365.5310000000027,
      "seconds_per_success": 11.076696969697052,
      "model_calls": 72
    },
    "lobster": {
      "episodes": 33,
      "pass": 33,
      "fail": 0,
      "error": 0,
      "success_rate": 1.0,
      "unsafe": 0,
      "unknown_safety": 0,
      "derived_usd": 0.9013002,
      "usd_per_success": 0.027312127272727273,
      "elapsed_s": 348.34399999998277,
      "seconds_per_success": 10.555878787878266,
      "model_calls": 72
    },
    "langgraph_parser": {
      "episodes": 33,
      "pass": 33,
      "fail": 0,
      "error": 0,
      "success_rate": 1.0,
      "unsafe": 0,
      "unknown_safety": 0,
      "derived_usd": 0.4927162,
      "usd_per_success": 0.014930793939393939,
      "elapsed_s": 205.34099999999307,
      "seconds_per_success": 6.222454545454336,
      "model_calls": 41
    },
    "lobster_parser": {
      "episodes": 33,
      "pass": 33,
      "fail": 0,
      "error": 0,
      "success_rate": 1.0,
      "unsafe": 0,
      "unknown_safety": 0,
      "derived_usd": 0.48845099999999997,
      "usd_per_success": 0.014801545454545454,
      "elapsed_s": 230.9269999999815,
      "seconds_per_success": 6.997787878787318,
      "model_calls": 40
    }
  },
  "failures": [
    {
      "arm": "plain",
      "task": "tb3_burger_conversation",
      "outcome": "FAIL",
      "reasons": [
        "missing_or_out_of_order_waypoint",
        "wrong_final_pose",
        "wrong_total_distance",
        "wrong_total_rotation"
      ],
      "error": null
    }
  ],
  "verification": {
    "complete": true,
    "rows": 198,
    "planned": 198,
    "rows_digest_matches": true,
    "lock_digest_matches": true,
    "source_snapshots_match": true,
    "regrade_available": true,
    "regrade_mismatches": [],
    "accounting_mismatches": [],
    "integrity": {
      "unchanged": true,
      "changed": [],
      "dependencies_unchanged": true
    },
    "mode": "live",
    "suite_split": "development"
  },
  "product_fixture_hashes_match": 166,
  "keys_scanned": 2,
  "journal_reconciles": true,
  "rows_sha256": "b7a5227a3abae39eca41d9f0ba254b96674d089bd80590e49b2f787e1a6ec2a4",
  "journal_sha256": "6df47151781b9478001350d05224054b70b2a537dc2bb3d5e44fa4b463293dc5",
  "comparisons": {
    "plain": {
      "pairs": 33,
      "independent_families": 19,
      "point": {
        "success_difference_pp": 3.0303030303030276,
        "cost_per_success_ratio": 0.5496858914603007,
        "time_per_success_ratio": 0.581300205625146
      },
      "family_bootstrap_95": {
        "success_difference_pp": [
          0.0,
          9.677419354838712
        ],
        "cost_per_success_ratio": [
          0.31417223480189505,
          0.8321581298473248
        ],
        "time_per_success_ratio": [
          0.36372827967878163,
          0.8313167092473612
        ]
      },
      "valid_bootstrap_draws": {
        "success_difference_pp": 2000,
        "cost_per_success_ratio": 2000,
        "time_per_success_ratio": 2000
      }
    },
    "langgraph": {
      "pairs": 33,
      "independent_families": 19,
      "point": {
        "success_difference_pp": 0.0,
        "cost_per_success_ratio": 0.5633853157125863,
        "time_per_success_ratio": 0.558776136634118
      },
      "family_bootstrap_95": {
        "success_difference_pp": [
          0.0,
          0.0
        ],
        "cost_per_success_ratio": [
          0.3209849040713661,
          0.8750063170714089
        ],
        "time_per_success_ratio": [
          0.3589289313500582,
          0.8015897424428754
        ]
      },
      "valid_bootstrap_draws": {
        "success_difference_pp": 2000,
        "cost_per_success_ratio": 2000,
        "time_per_success_ratio": 2000
      }
    },
    "lobster": {
      "pairs": 33,
      "independent_families": 19,
      "point": {
        "success_difference_pp": 0.0,
        "cost_per_success_ratio": 0.5529991006326194,
        "time_per_success_ratio": 0.5863456812806231
      },
      "family_bootstrap_95": {
        "success_difference_pp": [
          0.0,
          0.0
        ],
        "cost_per_success_ratio": [
          0.3143049743429368,
          0.8594180678280255
        ],
        "time_per_success_ratio": [
          0.36491319947412876,
          0.8651058513978196
        ]
      },
      "valid_bootstrap_draws": {
        "success_difference_pp": 2000,
        "cost_per_success_ratio": 2000,
        "time_per_success_ratio": 2000
      }
    },
    "langgraph_parser": {
      "pairs": 33,
      "independent_families": 19,
      "point": {
        "success_difference_pp": 0.0,
        "cost_per_success_ratio": 1.0115725847861303,
        "time_per_success_ratio": 0.9946868866909879
      },
      "family_bootstrap_95": {
        "success_difference_pp": [
          0.0,
          0.0
        ],
        "cost_per_success_ratio": [
          0.9425401173371424,
          1.1062224915401744
        ],
        "time_per_success_ratio": [
          0.9152832337898263,
          1.097641336394717
        ]
      },
      "valid_bootstrap_draws": {
        "success_difference_pp": 2000,
        "cost_per_success_ratio": 2000,
        "time_per_success_ratio": 2000
      }
    },
    "lobster_parser": {
      "pairs": 33,
      "independent_families": 19,
      "point": {
        "success_difference_pp": 0.0,
        "cost_per_success_ratio": 1.0204057315882247,
        "time_per_success_ratio": 0.8844786447666303
      },
      "family_bootstrap_95": {
        "success_difference_pp": [
          0.0,
          0.0
        ],
        "cost_per_success_ratio": [
          0.9830801524885687,
          1.0783230056108843
        ],
        "time_per_success_ratio": [
          0.7826251939784241,
          0.9927296559925542
        ]
      },
      "valid_bootstrap_draws": {
        "success_difference_pp": 2000,
        "cost_per_success_ratio": 2000,
        "time_per_success_ratio": 2000
      }
    }
  },
  "leadership_claim_allowed": false,
  "display_labels": {
    "omnilink_parser": "OmniLink",
    "plain": "Plain model loop",
    "langgraph": "LangGraph",
    "lobster": "Lobster SDK",
    "langgraph_parser": "LangGraph + OmniLink command handling",
    "lobster_parser": "Lobster SDK + OmniLink command handling"
  },
  "public_cost_scope": "Provider list-rate model costs, not invoice or total operating cost. Per-configuration costs exclude $0.006072 shared setup.",
  "reproduction_status": "Offline verification passed. Fresh live reproduction requires the matching runtime/source release; availability has not been verified."
}
