{
  "schema_version": "closureops.swe_marathon_evidence.v2",
  "generated_at": "2026-07-22T06:09:49Z",
  "campaign_hash": "sha256:2b10e08492f8da8ff20722a6f6c0e5864d13aebeebbeca50ac7e2fc104b7859a",
  "benchmark_source_commit": "6d6855af390226f6eca607d63818fe076e57ea8c",
  "harbor_version": "0.17.1",
  "selected_track": "canary",
  "expected_trial_count": 4,
  "observed_trial_count": 4,
  "campaign_results_complete": true,
  "complete_official_20_task_cohort": false,
  "provider_summaries": [
    {
      "provider_id": "qwen",
      "model": "dashscope/qwen3-coder-plus-2025-09-23",
      "trials": 2,
      "resolved_trials": 0,
      "resolved_rate": 0.0,
      "missing_rewards": 0,
      "errors": 0,
      "token_reporting_complete": true,
      "cost_reporting_complete": false,
      "input_tokens": 255769,
      "cache_tokens": 0,
      "output_tokens": 14151,
      "cost_usd": null
    },
    {
      "provider_id": "deepseek",
      "model": "deepseek/deepseek-v4-pro",
      "trials": 2,
      "resolved_trials": 0,
      "resolved_rate": 0.0,
      "missing_rewards": 0,
      "errors": 0,
      "token_reporting_complete": true,
      "cost_reporting_complete": true,
      "input_tokens": 429707,
      "cache_tokens": 381440,
      "output_tokens": 53684,
      "cost_usd": 0.06908394500000001
    }
  ],
  "trials": [
    {
      "provider_id": "deepseek",
      "model": "deepseek/deepseek-v4-pro",
      "task_id": "zstd-decoder",
      "trial_name": "zstd-decoder__ekdFmik",
      "task_checksum": "092ad3b8e200cf75a07d731feaa3348011417083f3b7b0f70429091b02ccdb0c",
      "resolved": false,
      "reward": 0,
      "missing_reward": false,
      "exception_type": "",
      "input_tokens": 65880,
      "cache_tokens": 50432,
      "output_tokens": 12198,
      "cost_usd": 0.017514956,
      "condition_id": "akrm_closure",
      "akrm_audit": {
        "required": true,
        "verified": true,
        "manifest_sha256": "sha256:ea005d2243962db973a4b1d75484e5fbc906efe0f561108d2a0e4540463ae678",
        "trace_sha256": "sha256:621aad59e8725f27405c805e998ed4ac8fe763e0a5425a0a038e8de4409d0952",
        "trace_records": 21,
        "chain_head": "sha256:5f75cb9a4fcfb608d07b7b786c1dd1f860368d605a6a9e7de1d00bdf9311a6f0",
        "max_statepack_tokens": 1015,
        "max_working_atoms": 44,
        "high_level_llm_calls": 20,
        "provider_request_attempts": 20,
        "output_length_retries": 0,
        "provider_usage_metering_complete": true,
        "provider_usage_trace_sha256": "sha256:533cdb268b1e857e7e5342a5e15f1e5fd20e5ac6e7216afee5144d2e28bc31e0",
        "provider_input_tokens_total": 65880,
        "provider_cache_tokens_total": 50432,
        "provider_output_tokens_total": 12198,
        "provider_max_input_tokens_per_response": 4477,
        "provider_cost_reporting_complete": true,
        "provider_cost_usd_total": 0.017514956
      },
      "baseline_metering_audit": {
        "required": false,
        "verified": false,
        "manifest_sha256": "",
        "high_level_llm_calls": 0,
        "provider_request_attempts": 0,
        "output_length_retries": 0,
        "provider_usage_metering_complete": false,
        "provider_usage_trace_sha256": "",
        "provider_input_tokens_total": 0,
        "provider_cache_tokens_total": 0,
        "provider_output_tokens_total": 0,
        "provider_max_input_tokens_per_response": 0,
        "provider_cost_reporting_complete": false,
        "provider_cost_usd_total": null
      },
      "attempt": 1
    },
    {
      "provider_id": "deepseek",
      "model": "deepseek/deepseek-v4-pro",
      "task_id": "zstd-decoder",
      "trial_name": "zstd-decoder__nDq5j2N",
      "task_checksum": "092ad3b8e200cf75a07d731feaa3348011417083f3b7b0f70429091b02ccdb0c",
      "resolved": false,
      "reward": 0,
      "missing_reward": false,
      "exception_type": "",
      "input_tokens": 363827,
      "cache_tokens": 331008,
      "output_tokens": 41486,
      "cost_usd": 0.051568989,
      "condition_id": "baseline_terminus2",
      "akrm_audit": {
        "required": false,
        "verified": false,
        "manifest_sha256": "",
        "trace_sha256": "",
        "trace_records": 0,
        "chain_head": "",
        "max_statepack_tokens": 0,
        "max_working_atoms": 0,
        "high_level_llm_calls": 0,
        "provider_request_attempts": 0,
        "output_length_retries": 0,
        "provider_usage_metering_complete": false,
        "provider_usage_trace_sha256": "",
        "provider_input_tokens_total": 0,
        "provider_cache_tokens_total": 0,
        "provider_output_tokens_total": 0,
        "provider_max_input_tokens_per_response": 0,
        "provider_cost_reporting_complete": false,
        "provider_cost_usd_total": null
      },
      "baseline_metering_audit": {
        "required": true,
        "verified": true,
        "manifest_sha256": "sha256:5e4171ca90af5dbd388128d70882ac3c417c5d6860d37f55b4b738c325c19683",
        "high_level_llm_calls": 20,
        "provider_request_attempts": 23,
        "output_length_retries": 3,
        "provider_usage_metering_complete": true,
        "provider_usage_trace_sha256": "sha256:e35bca83e97521e1ffa096fc8769a7abec82792af520378bef914d4958f75ec5",
        "provider_input_tokens_total": 363827,
        "provider_cache_tokens_total": 331008,
        "provider_output_tokens_total": 41486,
        "provider_max_input_tokens_per_response": 32676,
        "provider_cost_reporting_complete": true,
        "provider_cost_usd_total": 0.051568989
      },
      "attempt": 1
    },
    {
      "provider_id": "qwen",
      "model": "dashscope/qwen3-coder-plus-2025-09-23",
      "task_id": "zstd-decoder",
      "trial_name": "zstd-decoder__hPdMGUm",
      "task_checksum": "092ad3b8e200cf75a07d731feaa3348011417083f3b7b0f70429091b02ccdb0c",
      "resolved": false,
      "reward": 0,
      "missing_reward": false,
      "exception_type": "",
      "input_tokens": 56242,
      "cache_tokens": 0,
      "output_tokens": 2612,
      "cost_usd": null,
      "condition_id": "akrm_closure",
      "akrm_audit": {
        "required": true,
        "verified": true,
        "manifest_sha256": "sha256:dbdb1dd53daa7060cdf8af8ab77a37c59a195a8df7e1ae621afa05201c8f2733",
        "trace_sha256": "sha256:2a2e2259142f6d88ebae94c90c187911ab798143a858105f0ffb006eafdd64cb",
        "trace_records": 21,
        "chain_head": "sha256:9d06637f934c67aaa3960f8fc74ffcdec017c95846c472f331037e956732ae04",
        "max_statepack_tokens": 980,
        "max_working_atoms": 44,
        "high_level_llm_calls": 20,
        "provider_request_attempts": 20,
        "output_length_retries": 0,
        "provider_usage_metering_complete": true,
        "provider_usage_trace_sha256": "sha256:648e4cfdc55556c8d588ba29de9e1c9f65a240d2d2e4055429cc0a20efb3076b",
        "provider_input_tokens_total": 56242,
        "provider_cache_tokens_total": 0,
        "provider_output_tokens_total": 2612,
        "provider_max_input_tokens_per_response": 3596,
        "provider_cost_reporting_complete": false,
        "provider_cost_usd_total": null
      },
      "baseline_metering_audit": {
        "required": false,
        "verified": false,
        "manifest_sha256": "",
        "high_level_llm_calls": 0,
        "provider_request_attempts": 0,
        "output_length_retries": 0,
        "provider_usage_metering_complete": false,
        "provider_usage_trace_sha256": "",
        "provider_input_tokens_total": 0,
        "provider_cache_tokens_total": 0,
        "provider_output_tokens_total": 0,
        "provider_max_input_tokens_per_response": 0,
        "provider_cost_reporting_complete": false,
        "provider_cost_usd_total": null
      },
      "attempt": 1
    },
    {
      "provider_id": "qwen",
      "model": "dashscope/qwen3-coder-plus-2025-09-23",
      "task_id": "zstd-decoder",
      "trial_name": "zstd-decoder__jnarqf9",
      "task_checksum": "092ad3b8e200cf75a07d731feaa3348011417083f3b7b0f70429091b02ccdb0c",
      "resolved": false,
      "reward": 0,
      "missing_reward": false,
      "exception_type": "",
      "input_tokens": 199527,
      "cache_tokens": 0,
      "output_tokens": 11539,
      "cost_usd": null,
      "condition_id": "baseline_terminus2",
      "akrm_audit": {
        "required": false,
        "verified": false,
        "manifest_sha256": "",
        "trace_sha256": "",
        "trace_records": 0,
        "chain_head": "",
        "max_statepack_tokens": 0,
        "max_working_atoms": 0,
        "high_level_llm_calls": 0,
        "provider_request_attempts": 0,
        "output_length_retries": 0,
        "provider_usage_metering_complete": false,
        "provider_usage_trace_sha256": "",
        "provider_input_tokens_total": 0,
        "provider_cache_tokens_total": 0,
        "provider_output_tokens_total": 0,
        "provider_max_input_tokens_per_response": 0,
        "provider_cost_reporting_complete": false,
        "provider_cost_usd_total": null
      },
      "baseline_metering_audit": {
        "required": true,
        "verified": true,
        "manifest_sha256": "sha256:099ad59bfb1d1d15fecf065ef45b286d3b047473fb0378ebf0b5cc13e7e899de",
        "high_level_llm_calls": 20,
        "provider_request_attempts": 20,
        "output_length_retries": 0,
        "provider_usage_metering_complete": true,
        "provider_usage_trace_sha256": "sha256:d2dac2f36eff6d3b220cf810cabb8965c80678b227ea317986dc1df04b5b9b5c",
        "provider_input_tokens_total": 199527,
        "provider_cache_tokens_total": 0,
        "provider_output_tokens_total": 11539,
        "provider_max_input_tokens_per_response": 25161,
        "provider_cost_reporting_complete": false,
        "provider_cost_usd_total": null
      },
      "attempt": 1
    }
  ],
  "violations": [],
  "claim_boundary": {
    "public_engineering_benchmark_only": true,
    "gate_b_customer_evidence_authorized": false,
    "production_or_roi_claim_authorized": false,
    "independent_anti_cheat_review_completed": false,
    "official_leaderboard_submission_verified": false
  },
  "evidence_hash": "sha256:c02b6f8d70c9e06ee3529c4c24fc7e91443f89143c41a5058a17f9dc0b684fa6",
  "condition_summaries": [
    {
      "provider_id": "qwen",
      "condition_id": "baseline_terminus2",
      "model": "dashscope/qwen3-coder-plus-2025-09-23",
      "trials": 1,
      "resolved_trials": 0,
      "resolved_rate": 0.0,
      "missing_rewards": 0,
      "errors": 0,
      "token_reporting_complete": true,
      "cost_reporting_complete": false,
      "input_tokens": 199527,
      "cache_tokens": 0,
      "output_tokens": 11539,
      "cost_usd": null,
      "akrm_audits_required": 0,
      "akrm_audits_verified": 0
    },
    {
      "provider_id": "qwen",
      "condition_id": "akrm_closure",
      "model": "dashscope/qwen3-coder-plus-2025-09-23",
      "trials": 1,
      "resolved_trials": 0,
      "resolved_rate": 0.0,
      "missing_rewards": 0,
      "errors": 0,
      "token_reporting_complete": true,
      "cost_reporting_complete": false,
      "input_tokens": 56242,
      "cache_tokens": 0,
      "output_tokens": 2612,
      "cost_usd": null,
      "akrm_audits_required": 1,
      "akrm_audits_verified": 1
    },
    {
      "provider_id": "deepseek",
      "condition_id": "baseline_terminus2",
      "model": "deepseek/deepseek-v4-pro",
      "trials": 1,
      "resolved_trials": 0,
      "resolved_rate": 0.0,
      "missing_rewards": 0,
      "errors": 0,
      "token_reporting_complete": true,
      "cost_reporting_complete": true,
      "input_tokens": 363827,
      "cache_tokens": 331008,
      "output_tokens": 41486,
      "cost_usd": 0.051568989,
      "akrm_audits_required": 0,
      "akrm_audits_verified": 0
    },
    {
      "provider_id": "deepseek",
      "condition_id": "akrm_closure",
      "model": "deepseek/deepseek-v4-pro",
      "trials": 1,
      "resolved_trials": 0,
      "resolved_rate": 0.0,
      "missing_rewards": 0,
      "errors": 0,
      "token_reporting_complete": true,
      "cost_reporting_complete": true,
      "input_tokens": 65880,
      "cache_tokens": 50432,
      "output_tokens": 12198,
      "cost_usd": 0.017514956,
      "akrm_audits_required": 1,
      "akrm_audits_verified": 1
    }
  ],
  "paired_comparisons": [
    {
      "provider_id": "qwen",
      "model": "dashscope/qwen3-coder-plus-2025-09-23",
      "pairs": 1,
      "both_resolved": 0,
      "akrm_only_resolved": 0,
      "baseline_only_resolved": 0,
      "neither_resolved": 1,
      "akrm_minus_baseline_resolved_rate": 0.0,
      "exact_mcnemar_two_sided_p": 1.0,
      "baseline_input_tokens": 199527,
      "akrm_input_tokens": 56242,
      "input_token_reduction_rate": 0.71812336175054
    },
    {
      "provider_id": "deepseek",
      "model": "deepseek/deepseek-v4-pro",
      "pairs": 1,
      "both_resolved": 0,
      "akrm_only_resolved": 0,
      "baseline_only_resolved": 0,
      "neither_resolved": 1,
      "akrm_minus_baseline_resolved_rate": 0.0,
      "exact_mcnemar_two_sided_p": 1.0,
      "baseline_input_tokens": 363827,
      "akrm_input_tokens": 65880,
      "input_token_reduction_rate": 0.8189249286061782
    }
  ],
  "paired_fairness_verified": true
}
