{
  "task_set_version": "0.19.0",
  "scope": "wave19 tasks432–471; historical composites explicitly labeled",
  "baseline_wave19": {
    "wave19-giga35": {
      "passed": 0,
      "total": 40,
      "wilson95": [
        6.938893903907228e-18,
        0.08762160119728224
      ]
    },
    "wave19-haiku": {
      "passed": 31,
      "total": 40,
      "wilson95": [
        0.6249690325769044,
        0.8768390867645905
      ]
    },
    "wave19-glm53": {
      "passed": 9,
      "total": 40,
      "wilson95": [
        0.12316091323540954,
        0.3750309674230957
      ]
    },
    "wave19-deepseek41flash": {
      "passed": 27,
      "total": 40,
      "wilson95": [
        0.5201774618987811,
        0.7991549776821703
      ]
    }
  },
  "rows": [
    {
      "id": "w19-codex-giga35",
      "harness": "Codex CLI 0.160.1",
      "model": "GigaChat 3.5 (PROM), build 32.9.16.9 (default)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 0,
        "total": 40,
        "wilson95": [
          6.938893903907228e-18,
          0.08762160119728224
        ]
      },
      "source_sha256": "0fc6528cfb7f3dc59218854d74e49c364598bdb89ee889d0baa4b6d3146da64e",
      "reasoning_effort": "default",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-codex-giga35",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-codex-haiku",
      "harness": "Codex CLI 0.160.1",
      "model": "Claude Haiku 5.5 (OpenRouter) (default)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 31,
        "total": 40,
        "wilson95": [
          0.6249690325769044,
          0.8768390867645905
        ]
      },
      "source_sha256": "71e30121e573f90983ddbc9bcef3bf887b3dfb5b65d7e47252957a4bf407c0d3",
      "reasoning_effort": "default",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-codex-haiku",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-codex-glm53",
      "harness": "Codex CLI 0.160.1",
      "model": "GLM-5.3 (OpenRouter) (high)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 14,
        "total": 40,
        "wilson95": [
          0.22134528873176587,
          0.5049411916274189
        ]
      },
      "source_sha256": "bb02337d2b7ed2980d74893e78300217d165d51b80e840e2e1ff316e3e954e14",
      "reasoning_effort": "high",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-codex-glm53",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-codex-deepseek41flash",
      "harness": "Codex CLI 0.160.1",
      "model": "DeepSeek V4.1 Flash (OpenRouter) (high)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 31,
        "total": 40,
        "wilson95": [
          0.6249690325769044,
          0.8768390867645905
        ]
      },
      "source_sha256": "c08925f32471f8b8b50a0c3b11f530b49df62795d78a47c33e31da13ed60d297",
      "reasoning_effort": "high",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-codex-deepseek41flash",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-claude-giga35",
      "harness": "Claude Code 2.1.300",
      "model": "GigaChat 3.5 (PROM), build 32.9.16.9 (default)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 2,
        "total": 40,
        "wilson95": [
          0.013820667386148802,
          0.16503877369140524
        ]
      },
      "source_sha256": "1ad88b0a038f3527fa3775375e07e8f1ef410367b8e73592a95fa79c0203089f",
      "reasoning_effort": "default",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-claude-giga35",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-claude-haiku",
      "harness": "Claude Code 2.1.300",
      "model": "Claude Haiku 5.5 (OpenRouter) (default)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 26,
        "total": 40,
        "wilson95": [
          0.4950588083725812,
          0.7786547112682342
        ]
      },
      "source_sha256": "562ededc1c66362b0cc5672bc3e0aab90898acb8ceb6d4acf0627e163af3f8c2",
      "reasoning_effort": "default",
      "measurement_date": "2026-10-20",
      "details_url": "harness-wave19.html#w19-claude-haiku",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-claude-glm53",
      "harness": "Claude Code 2.1.300",
      "model": "GLM-5.3 (OpenRouter) (high)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 17,
        "total": 40,
        "wilson95": [
          0.2850939255995071,
          0.5780493145800853
        ]
      },
      "source_sha256": "b3bed91d409f76fc215d19c2b417ee5166d46d4ae1a3578fb164df78a5f3e9b7",
      "reasoning_effort": "high",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-claude-glm53",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-claude-deepseek41flash",
      "harness": "Claude Code 2.1.300",
      "model": "DeepSeek V4.1 Flash (OpenRouter) (high)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 30,
        "total": 40,
        "wilson95": [
          0.5980603857923241,
          0.8581288136090348
        ]
      },
      "source_sha256": "0795af44ce39fe693fd679dd25146336e967ef89be8a1f6ec3cdefd2f898690c",
      "reasoning_effort": "high",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-claude-deepseek41flash",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-pi-giga35",
      "harness": "pi-mono 0.73.1",
      "model": "GigaChat 3.5 (PROM), build 32.9.16.9 (default)",
      "runs": 1,
      "source_version": "0.18.0",
      "measurement_type": "historical_plus_wave19",
      "source_passed": 294,
      "light": {
        "passed": 294,
        "total": 391,
        "wilson95": [
          0.7067958007003114,
          0.7921386338658973
        ]
      },
      "medium": {
        "passed": 0,
        "total": 80,
        "wilson95": [
          3.469446951953614e-18,
          0.04581812953552469
        ]
      },
      "medium_historical": {
        "passed": 0,
        "total": 40,
        "wilson95": [
          6.938893903907228e-18,
          0.08762160119728224
        ]
      },
      "wave19": {
        "passed": 0,
        "total": 40,
        "wilson95": [
          6.938893903907228e-18,
          0.08762160119728224
        ]
      },
      "source_sha256": "c180d6747bc518c70fba134a5f46daba5d986ec98636aa492d0db510dcd4f0c0",
      "reasoning_effort": "default",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-pi-giga35",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures · historical431 + new40, not a fresh full run",
      "historical_provenance": {
        "source_version": "0.18.0",
        "source_sha256": "67e5336c70fe40e65f31b05710e4fb08d24f71a33910908ad229e7fae2bd3c36",
        "validation": "raw_431_verified",
        "light_passed": 294,
        "medium40_passed": 0
      },
      "extends_id": "run-19"
    },
    {
      "id": "w19-pi-haiku",
      "harness": "pi-mono 0.73.1",
      "model": "Claude Haiku 5.5 (OpenRouter) (default)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 29,
        "total": 40,
        "wilson95": [
          0.5716504419378987,
          0.8389198375233242
        ]
      },
      "source_sha256": "f85833c14926faeed84c35f9aa748ab0418270f521a9cfa880ea6ea279b0a71d",
      "reasoning_effort": "default",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-pi-haiku",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-pi-glm53",
      "harness": "pi-mono 0.73.1",
      "model": "GLM-5.3 (OpenRouter) (high)",
      "runs": 1,
      "source_version": "0.18.0",
      "measurement_type": "historical_plus_wave19",
      "source_passed": 396,
      "light": {
        "passed": 376,
        "total": 391,
        "wilson95": [
          0.9376751972598407,
          0.9766158224274761
        ]
      },
      "medium": {
        "passed": 44,
        "total": 80,
        "wilson95": [
          0.4411950788935626,
          0.6542231081528851
        ]
      },
      "medium_historical": {
        "passed": 20,
        "total": 40,
        "wilson95": [
          0.3519952693346575,
          0.6480047306653425
        ]
      },
      "wave19": {
        "passed": 24,
        "total": 40,
        "wilson95": [
          0.44595893660346597,
          0.7365167431570776
        ]
      },
      "source_sha256": "fa550a21dc95c1873b93be198f1f3471075a123bfa0bdd1d3337c30c2a2ea061",
      "reasoning_effort": "high",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-pi-glm53",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures · historical431 + new40, not a fresh full run",
      "historical_provenance": {
        "source_version": "0.18.0",
        "source_sha256": null,
        "validation": "published_aggregate",
        "light_passed": 376,
        "medium40_passed": 20
      },
      "extends_id": "glm53-openrouter-pi"
    },
    {
      "id": "w19-pi-deepseek41flash",
      "harness": "pi-mono 0.73.1",
      "model": "DeepSeek V4.1 Flash (OpenRouter) (high)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 26,
        "total": 40,
        "wilson95": [
          0.4950588083725812,
          0.7786547112682342
        ]
      },
      "source_sha256": "3db937aeb3d715e2d9a0348bab5d08a8a5c522bee730353fe4a00380920b83da",
      "reasoning_effort": "high",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-pi-deepseek41flash",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-openclaw-giga35",
      "harness": "OpenClaw 2026.9.6",
      "model": "GigaChat 3.5 (PROM), build 32.9.16.9 (default)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 0,
        "total": 40,
        "wilson95": [
          6.938893903907228e-18,
          0.08762160119728224
        ]
      },
      "source_sha256": "ab9c64f4130b90ad8b3dba4130efca199bf2390fa72decd9a21551dad20bc51c",
      "reasoning_effort": "default",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-openclaw-giga35",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-openclaw-haiku",
      "harness": "OpenClaw 2026.9.6",
      "model": "Claude Haiku 5.5 (OpenRouter) (default)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 23,
        "total": 40,
        "wilson95": [
          0.42195068541991465,
          0.7149060744004929
        ]
      },
      "source_sha256": "639e0aaad318d7e1c3cab7b4d5b66795a61e42c4414bc1be61d715a82bca8a3d",
      "reasoning_effort": "default",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-openclaw-haiku",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    },
    {
      "id": "w19-openclaw-glm53",
      "harness": "OpenClaw 2026.9.6",
      "model": "GLM-5.3 (OpenRouter) (high)",
      "runs": 1,
      "source_version": "0.18.0",
      "measurement_type": "historical_plus_wave19",
      "source_passed": 408,
      "light": {
        "passed": 385,
        "total": 391,
        "wilson95": [
          0.9669302938675344,
          0.9929486436989285
        ]
      },
      "medium": {
        "passed": 41,
        "total": 80,
        "wilson95": [
          0.40493334216542537,
          0.6189212045961865
        ]
      },
      "medium_historical": {
        "passed": 23,
        "total": 40,
        "wilson95": [
          0.42195068541991465,
          0.7149060744004929
        ]
      },
      "wave19": {
        "passed": 18,
        "total": 40,
        "wilson95": [
          0.30705307810905275,
          0.6017090820106756
        ]
      },
      "source_sha256": "4c35e9cf435d6f53a0d768870b34007c234c102d5436563277c9d9afeaca8a49",
      "reasoning_effort": "high",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-openclaw-glm53",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures · historical431 + new40, not a fresh full run",
      "historical_provenance": {
        "source_version": "0.18.0",
        "source_sha256": "0f8edf1e9af3e9423e982a7a4c74a1aa0d0f92c9ed1bbcac89592127ad731936",
        "validation": "raw_431_verified",
        "light_passed": 385,
        "medium40_passed": 23
      }
    },
    {
      "id": "w19-openclaw-deepseek41flash",
      "harness": "OpenClaw 2026.9.6",
      "model": "DeepSeek V4.1 Flash (OpenRouter) (high)",
      "runs": 1,
      "source_version": "0.19.0",
      "measurement_type": "wave19_only",
      "source_passed": null,
      "light": null,
      "medium": null,
      "medium_historical": null,
      "wave19": {
        "passed": 28,
        "total": 40,
        "wilson95": [
          0.545699811818555,
          0.8192515477025322
        ]
      },
      "source_sha256": "ff342b61d477f048db68e78756c5bd076b8ea75657db5e845e5d574ea4b88494",
      "reasoning_effort": "high",
      "measurement_date": "2026-10-19",
      "details_url": "harness-wave19.html#w19-openclaw-deepseek41flash",
      "telemetry_note": "Wave 19 only · tasks432–471 · k=1 · 120 seconds · zero final infrastructure failures"
    }
  ],
  "audit": {
    "coverage": "640 unique configuration/task outcomes; each configuration has40 unique current tasks",
    "terminal_api_errors": "Three ClaudeCode/Haiku invalid-request failures repaired after confirmed adapter ordering bug; other37 retained unchanged. Final nativeinfra0.",
    "repair_tasks": [
      "task_452_search_hint",
      "task_461_search_pool",
      "task_465_search_count"
    ],
    "original_haiku_passed": 23,
    "corrected_haiku_passed": 26,
    "limitation": "Coverage, runtime protocol and terminal API error screening; not an exhaustive qualitative study of every failure."
  }
}