{
  "schemaVersion": 1,
  "benchmark": {
    "datasetRef": "4",
    "name": "Terminal-Bench",
    "score": "accuracy",
    "scoreUnit": "percent",
    "taskCount": 66,
    "trialsPerConfiguration": 330,
    "trialsPerTask": 5,
    "version": "4.0.0",
    "runPolicy": {
      "actionLimit": null,
      "agentCount": null,
      "agentTimeoutSeconds": 28800,
      "costBasis": "source-reported-total-usd; pricing basis unspecified",
      "disqualifiedTrialPolicy": "reject-submission",
      "retryPolicy": null,
      "seedPolicy": null,
      "tokenLimit": null,
      "toolsMode": null,
      "uncertaintyMethod": "source-reported-95%-confidence-interval; method unspecified"
    }
  },
  "source": {
    "leaderboardDefinitionUrl": "https://github.com/harbor-framework/terminal-bench/blob/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/leaderboard.yaml",
    "leaderboardUrl": "https://hub.harborframework.com/datasets/terminal-bench/terminal-bench/latest?leaderboard=4-0-0&tab=leaderboard",
    "method": "version-pinned-github-submissions",
    "name": "Harbor Framework",
    "releaseUrl": "https://github.com/harbor-framework/terminal-bench/releases/tag/v4.0.0",
    "repositoryCommit": "3e9bf0ee42f7419722428680966435a7a71d0d40",
    "repositoryCommittedAt": "2026-10-06T23:19:56Z",
    "repositoryCommitUrl": "https://github.com/harbor-framework/terminal-bench/commit/3e9bf0ee42f7419722428680966435a7a71d0d40",
    "repositoryUrl": "https://github.com/harbor-framework/terminal-bench",
    "retrievedAt": "2026-10-09T16:00:31.374Z",
    "sourceClass": "benchmark-owner",
    "submissionsDirectoryUrl": "https://github.com/harbor-framework/terminal-bench/tree/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions"
  },
  "records": [
    {
      "id": "2026-09-30-anthropic-claude-opus-5-5-max-claude-code",
      "releaseDate": "2026-09-22",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.284"
      },
      "model": {
        "display": {
          "label": "Opus 5.5",
          "url": "https://www.anthropic.com/claude-opus-5-5"
        },
        "id": "anthropic/claude-opus-5-5",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        }
      },
      "metrics": {
        "accuracyPercent": 64.85,
        "accuracyCi95HalfWidthPercent": 3.11,
        "averageTrialDurationSeconds": 4790.4,
        "cachedInputTokens": 7699681320,
        "nTrials": 330,
        "outputTokens": 88814607,
        "passAt2": 0.7318,
        "passAt3": 0.7652,
        "passAt4": 0.7788,
        "passAt5": 0.7879,
        "successes": 214,
        "totalCostUsd": 4718.65,
        "totalTokens": 8034082467,
        "uncachedInputTokens": 7945267860
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/bfd722a4-4c18-5a00-a874-c70b646870b8"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-30-anthropic-claude-opus-5-5-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-30-anthropic-claude-opus-5-5-max-claude-code.json"
    },
    {
      "id": "2026-09-30-anthropic-claude-sonnet-5-5-max-claude-code",
      "releaseDate": "2026-09-28",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.284"
      },
      "model": {
        "display": {
          "label": "Sonnet 5.5",
          "url": "https://www.anthropic.com/claude-sonnet-5-5"
        },
        "id": "anthropic/claude-sonnet-5-5",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        }
      },
      "metrics": {
        "accuracyPercent": 61.82,
        "accuracyCi95HalfWidthPercent": 2.94,
        "averageTrialDurationSeconds": 7431.6,
        "cachedInputTokens": 18580542667,
        "nTrials": 330,
        "outputTokens": 186577885,
        "passAt2": 0.6924,
        "passAt3": 0.7364,
        "passAt4": 0.7606,
        "passAt5": 0.7727,
        "successes": 204,
        "totalCostUsd": 7333.64,
        "totalTokens": 19413047384,
        "uncachedInputTokens": 19226469499
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/4e34616a-9b52-50b7-adef-136579456b21"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-30-anthropic-claude-sonnet-5-5-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-30-anthropic-claude-sonnet-5-5-max-claude-code.json"
    },
    {
      "id": "2026-09-30-openai-gpt-6-1-sol-max-codex",
      "releaseDate": "2026-09-29",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Codex",
          "url": "https://openai.com/codex/"
        },
        "id": "codex",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        },
        "version": "0.159.0"
      },
      "model": {
        "display": {
          "label": "GPT-6.1 Sol",
          "url": "https://developers.openai.com/api/docs/models/gpt-6.1-sol"
        },
        "id": "openai/gpt-6.1-sol",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        }
      },
      "metrics": {
        "accuracyPercent": 58.18,
        "accuracyCi95HalfWidthPercent": 3.11,
        "averageTrialDurationSeconds": 3176.7,
        "cachedInputTokens": 1354187811,
        "nTrials": 330,
        "outputTokens": 31314516,
        "passAt2": 0.6652,
        "passAt3": 0.7015,
        "passAt4": 0.7242,
        "passAt5": 0.7424,
        "successes": 192,
        "totalCostUsd": 634.24,
        "totalTokens": 1459936514,
        "uncachedInputTokens": 1428621998
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/3837153f-6673-539d-b3e2-99a6edd3c5db"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-30-openai-gpt-6-1-sol-max-codex.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-30-openai-gpt-6-1-sol-max-codex.json"
    },
    {
      "id": "2026-09-02-anthropic-claude-fable-5-1-max-claude-code",
      "releaseDate": "2026-09-01",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.257"
      },
      "model": {
        "display": {
          "label": "Fable 5.1",
          "url": "https://docs.anthropic.com/en/docs/about-claude/models/all-models"
        },
        "id": "anthropic/claude-fable-5-1",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        }
      },
      "metrics": {
        "accuracyPercent": 57.88,
        "accuracyCi95HalfWidthPercent": 3.76,
        "averageTrialDurationSeconds": 3893.5,
        "cachedInputTokens": 2467056906,
        "nTrials": 330,
        "outputTokens": 63121568,
        "passAt2": 0.7,
        "passAt3": 0.747,
        "passAt4": 0.7727,
        "passAt5": 0.7879,
        "successes": 191,
        "totalCostUsd": 6243.5,
        "totalTokens": 2746221560,
        "uncachedInputTokens": 2683099992
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/d3a44a19-9643-5d66-9633-0375f701d4db"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-02-anthropic-claude-fable-5-1-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-02-anthropic-claude-fable-5-1-max-claude-code.json"
    },
    {
      "id": "2026-08-26-anthropic-claude-opus-5-max-claude-code",
      "releaseDate": "2026-07-24",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.231"
      },
      "model": {
        "display": {
          "label": "Opus 5",
          "url": "https://www.anthropic.com/news/claude-opus-5"
        },
        "id": "anthropic/claude-opus-5",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        }
      },
      "metrics": {
        "accuracyPercent": 51.82,
        "accuracyCi95HalfWidthPercent": 3.39,
        "averageTrialDurationSeconds": 4792.8,
        "cachedInputTokens": 6271909956,
        "nTrials": 330,
        "outputTokens": 66012156,
        "passAt2": 0.6167,
        "passAt3": 0.6576,
        "passAt4": 0.6818,
        "passAt5": 0.697,
        "successes": 171,
        "totalCostUsd": 5969.11,
        "totalTokens": 6527147637,
        "uncachedInputTokens": 6461135481
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/a1ac63a1-8a9b-4bc7-9906-2b63657ee1c2"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-anthropic-claude-opus-5-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-anthropic-claude-opus-5-max-claude-code.json"
    },
    {
      "id": "2026-09-30-openai-gpt-6-sol-max-codex",
      "releaseDate": "2026-09-22",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Codex",
          "url": "https://openai.com/codex/"
        },
        "id": "codex",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        },
        "version": "0.159.0"
      },
      "model": {
        "display": {
          "label": "GPT-6 Sol",
          "url": "https://developers.openai.com/api/docs/models/gpt-6-sol"
        },
        "id": "openai/gpt-6-sol",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        }
      },
      "metrics": {
        "accuracyPercent": 49.39,
        "accuracyCi95HalfWidthPercent": 3.23,
        "averageTrialDurationSeconds": 2662.2,
        "cachedInputTokens": 4879399009,
        "nTrials": 330,
        "outputTokens": 30571836,
        "passAt2": 0.5833,
        "passAt3": 0.6242,
        "passAt4": 0.6485,
        "passAt5": 0.6667,
        "successes": 163,
        "totalCostUsd": 1451.32,
        "totalTokens": 4977940044,
        "uncachedInputTokens": 4947368208
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/67472e26-b1ec-5f01-9cd8-85e1188af1c8"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-30-openai-gpt-6-sol-max-codex.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-30-openai-gpt-6-sol-max-codex.json"
    },
    {
      "id": "2026-08-26-anthropic-claude-fable-5-max-claude-code",
      "releaseDate": "2026-06-09",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.231"
      },
      "model": {
        "display": {
          "label": "Fable 5",
          "url": "https://www.anthropic.com/news/claude-fable-5-mythos-5"
        },
        "id": "anthropic/claude-fable-5",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        }
      },
      "metrics": {
        "accuracyPercent": 44.55,
        "accuracyCi95HalfWidthPercent": 3.85,
        "averageTrialDurationSeconds": 4202.8,
        "cachedInputTokens": 3564065093,
        "nTrials": 330,
        "outputTokens": 58619905,
        "passAt2": 0.5727,
        "passAt3": 0.6318,
        "passAt4": 0.6636,
        "passAt5": 0.6818,
        "successes": 147,
        "totalCostUsd": 7265.01,
        "totalTokens": 3785220369,
        "uncachedInputTokens": 3726600464
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/991af5e0-8c5f-4511-9619-a2a46939ebd4"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-anthropic-claude-fable-5-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-anthropic-claude-fable-5-max-claude-code.json"
    },
    {
      "id": "2026-08-26-anthropic-glm-5-3-max-claude-code",
      "releaseDate": "2026-08-14",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.207"
      },
      "model": {
        "display": {
          "label": "GLM-5.3",
          "url": "https://docs.z.ai/guides/llm/glm-5.3"
        },
        "id": "anthropic/glm-5.3",
        "organization": {
          "label": "Z.ai",
          "url": "https://z.ai"
        }
      },
      "metrics": {
        "accuracyPercent": 41.82,
        "accuracyCi95HalfWidthPercent": 3.23,
        "averageTrialDurationSeconds": 5831,
        "cachedInputTokens": 8444041600,
        "nTrials": 330,
        "outputTokens": 68660047,
        "passAt2": 0.5076,
        "passAt3": 0.547,
        "passAt4": 0.5667,
        "passAt5": 0.5758,
        "successes": 138,
        "totalCostUsd": 2727.63,
        "totalTokens": 8677039373,
        "uncachedInputTokens": 8608379326
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/18e4ee29-d8b4-516b-b79d-b8bad5bcd5be"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-anthropic-glm-5-3-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-anthropic-glm-5-3-max-claude-code.json"
    },
    {
      "id": "2026-08-26-openai-gpt-5-6-sol-max-codex",
      "releaseDate": "2026-06-26",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Codex",
          "url": "https://openai.com/codex/"
        },
        "id": "codex",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        },
        "version": "0.149.1"
      },
      "model": {
        "display": {
          "label": "GPT-5.6 Sol",
          "url": "https://developers.openai.com/api/docs/models/gpt-5.6-sol"
        },
        "id": "openai/gpt-5.6-sol",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        }
      },
      "metrics": {
        "accuracyPercent": 37.27,
        "accuracyCi95HalfWidthPercent": 3.78,
        "averageTrialDurationSeconds": 2388.4,
        "cachedInputTokens": 4316454466,
        "nTrials": 330,
        "outputTokens": 23451415,
        "passAt2": 0.4955,
        "passAt3": 0.55,
        "passAt4": 0.5818,
        "passAt5": 0.6061,
        "successes": 123,
        "totalCostUsd": 2541.7,
        "totalTokens": 4409948969,
        "uncachedInputTokens": 4386497554
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/5aebbab5-1f8f-4cdb-8073-c6ce785a930d"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-openai-gpt-5-6-sol-max-codex.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-openai-gpt-5-6-sol-max-codex.json"
    },
    {
      "id": "2026-10-06-z-ai-glm-5-3-flash-none-claude-code",
      "releaseDate": "2026-08-26",
      "reasoningEffort": "none",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.285"
      },
      "model": {
        "display": {
          "label": "GLM-5.3-Flash",
          "url": "https://z.ai/blog/glm-5.3-flash"
        },
        "id": "z-ai/glm-5.3-flash",
        "organization": {
          "label": "Z.ai",
          "url": "https://z.ai"
        }
      },
      "metrics": {
        "accuracyPercent": 35.76,
        "accuracyCi95HalfWidthPercent": 3.54,
        "averageTrialDurationSeconds": 6209.8,
        "cachedInputTokens": 2786100736,
        "nTrials": 330,
        "outputTokens": 44554633,
        "passAt2": 0.4652,
        "passAt3": 0.5136,
        "passAt4": 0.5424,
        "passAt5": 0.5606,
        "successes": 118,
        "totalCostUsd": 2883.11,
        "totalTokens": 2916503275,
        "uncachedInputTokens": 2871948642
      },
      "sourceFilterReasoningEffort": null,
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/e498ef42-3647-58ea-8790-b18f53c18d89"
      ],
      "sourcePath": "leaderboard/submissions/2026-10-06-z-ai-glm-5-3-flash-none-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-10-06-z-ai-glm-5-3-flash-none-claude-code.json"
    },
    {
      "id": "2026-10-06-qwen-qwen3-8-max-0902-max-claude-code",
      "releaseDate": "2026-09-02",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.285"
      },
      "model": {
        "display": {
          "label": "Qwen3.8-Max-0902",
          "url": "https://www.alibabacloud.com/en/notice/model_studio_update_notice_for_qwen38max_models_863"
        },
        "id": "qwen/qwen3.8-max-0902",
        "organization": {
          "label": "Alibaba",
          "url": "https://www.alibabacloud.com"
        }
      },
      "metrics": {
        "accuracyPercent": 26.97,
        "accuracyCi95HalfWidthPercent": 3.78,
        "averageTrialDurationSeconds": 9953,
        "cachedInputTokens": 2702893266,
        "nTrials": 330,
        "outputTokens": 83393525,
        "passAt2": 0.3924,
        "passAt3": 0.4591,
        "passAt4": 0.5,
        "passAt5": 0.5303,
        "successes": 89,
        "totalCostUsd": 4635.67,
        "totalTokens": 2957967711,
        "uncachedInputTokens": 2874574186
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/e630e811-bfab-5823-8acc-0224c6e2dba9"
      ],
      "sourcePath": "leaderboard/submissions/2026-10-06-qwen-qwen3-8-max-0902-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-10-06-qwen-qwen3-8-max-0902-max-claude-code.json"
    },
    {
      "id": "2026-08-26-anthropic-claude-opus-4-8-max-claude-code",
      "releaseDate": "2026-05-28",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.231"
      },
      "model": {
        "display": {
          "label": "Opus 4.8",
          "url": "https://www.anthropic.com/news/claude-opus-4-8"
        },
        "id": "anthropic/claude-opus-4-8",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        }
      },
      "metrics": {
        "accuracyPercent": 23.64,
        "accuracyCi95HalfWidthPercent": 3.56,
        "averageTrialDurationSeconds": 5163,
        "cachedInputTokens": 6136275149,
        "nTrials": 330,
        "outputTokens": 88813309,
        "passAt2": 0.3455,
        "passAt3": 0.4045,
        "passAt4": 0.4424,
        "passAt5": 0.4697,
        "successes": 78,
        "totalCostUsd": 6481.26,
        "totalTokens": 6415823126,
        "uncachedInputTokens": 6327009817
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/10722b22-0bbd-4c93-8d27-611fb8dd02a6"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-anthropic-claude-opus-4-8-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-anthropic-claude-opus-4-8-max-claude-code.json"
    },
    {
      "id": "2026-08-26-openai-gpt-5-6-terra-max-codex",
      "releaseDate": "2026-06-26",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Codex",
          "url": "https://openai.com/codex/"
        },
        "id": "codex",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        },
        "version": "0.149.1"
      },
      "model": {
        "display": {
          "label": "GPT-5.6 Terra",
          "url": "https://developers.openai.com/api/docs/models/gpt-5.6-terra"
        },
        "id": "openai/gpt-5.6-terra",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        }
      },
      "metrics": {
        "accuracyPercent": 21.52,
        "accuracyCi95HalfWidthPercent": 3.25,
        "averageTrialDurationSeconds": 2514.6,
        "cachedInputTokens": 5560907776,
        "nTrials": 330,
        "outputTokens": 33224833,
        "passAt2": 0.3061,
        "passAt3": 0.3636,
        "passAt4": 0.4061,
        "passAt5": 0.4394,
        "successes": 71,
        "totalCostUsd": 1733.52,
        "totalTokens": 5683805798,
        "uncachedInputTokens": 5650580965
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/3de0aee9-6de7-4d94-83c4-ac5e789fe757"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-openai-gpt-5-6-terra-max-codex.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-openai-gpt-5-6-terra-max-codex.json"
    },
    {
      "id": "2026-08-26-xai-grok-4-6-none-grok-build",
      "releaseDate": "2026-08-12",
      "reasoningEffort": "none",
      "harness": {
        "display": {
          "label": "Grok Build",
          "url": "https://x.ai/build"
        },
        "id": "grok-build",
        "organization": {
          "label": "xAI",
          "url": "https://x.ai"
        },
        "version": "1.0.5"
      },
      "model": {
        "display": {
          "label": "Grok 4.6",
          "url": "https://docs.x.ai/developers/models/grok-4.6"
        },
        "id": "xai/grok-4.6",
        "organization": {
          "label": "xAI",
          "url": "https://x.ai"
        }
      },
      "metrics": {
        "accuracyPercent": 20.3,
        "accuracyCi95HalfWidthPercent": 3.09,
        "averageTrialDurationSeconds": 2351.5,
        "cachedInputTokens": 3877638528,
        "nTrials": 330,
        "outputTokens": 32618221,
        "passAt2": 0.2848,
        "passAt3": 0.3303,
        "passAt4": 0.3636,
        "passAt5": 0.3939,
        "successes": 67,
        "totalCostUsd": 3591.58,
        "totalTokens": 4001923428,
        "uncachedInputTokens": 3969305207
      },
      "sourceFilterReasoningEffort": null,
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/2c89a14f-d14a-4ea8-ad8a-d08d90c67a5d"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-xai-grok-4-6-none-grok-build.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-xai-grok-4-6-none-grok-build.json"
    },
    {
      "id": "2026-09-01-gemini-gemini-3-8-flash-high-mini-swe-agent",
      "releaseDate": "2026-09-02",
      "reasoningEffort": "high",
      "harness": {
        "display": {
          "label": "mini-SWE-agent",
          "url": "https://github.com/SWE-agent/mini-swe-agent"
        },
        "id": "mini-swe-agent",
        "organization": {
          "label": "SWE-agent",
          "url": "https://github.com/SWE-agent"
        },
        "version": "2.4.6"
      },
      "model": {
        "display": {
          "label": "Gemini 3.8 Flash",
          "url": "https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash"
        },
        "id": "gemini/gemini-3.8-flash",
        "organization": {
          "label": "Google",
          "url": "https://google.com"
        }
      },
      "metrics": {
        "accuracyPercent": 19.09,
        "accuracyCi95HalfWidthPercent": 3.36,
        "averageTrialDurationSeconds": 1995.1,
        "cachedInputTokens": 16692887408,
        "nTrials": 330,
        "outputTokens": 68056484,
        "passAt2": 0.2879,
        "passAt3": 0.353,
        "passAt4": 0.4,
        "passAt5": 0.4394,
        "successes": 63,
        "totalCostUsd": 0,
        "totalTokens": 17189729495,
        "uncachedInputTokens": 17121673011
      },
      "sourceFilterReasoningEffort": "high",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/542367e9-c4ae-5255-ab76-4c9045749e8e"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-01-gemini-gemini-3-8-flash-high-mini-swe-agent.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-01-gemini-gemini-3-8-flash-high-mini-swe-agent.json"
    },
    {
      "id": "2026-08-26-openai-gpt-5-6-luna-max-codex",
      "releaseDate": "2026-06-26",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Codex",
          "url": "https://openai.com/codex/"
        },
        "id": "codex",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        },
        "version": "0.149.1"
      },
      "model": {
        "display": {
          "label": "GPT-5.6 Luna",
          "url": "https://developers.openai.com/api/docs/models/gpt-5.6-luna"
        },
        "id": "openai/gpt-5.6-luna",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        }
      },
      "metrics": {
        "accuracyPercent": 17.27,
        "accuracyCi95HalfWidthPercent": 2.85,
        "averageTrialDurationSeconds": 4088.3,
        "cachedInputTokens": 11305212851,
        "nTrials": 330,
        "outputTokens": 61282672,
        "passAt2": 0.2424,
        "passAt3": 0.2848,
        "passAt4": 0.3121,
        "passAt5": 0.3333,
        "successes": 57,
        "totalCostUsd": 346.67,
        "totalTokens": 11557463640,
        "uncachedInputTokens": 11496180968
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/b7444ba2-78b7-46a0-a3c1-2196d40a337d"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-openai-gpt-5-6-luna-max-codex.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-openai-gpt-5-6-luna-max-codex.json"
    },
    {
      "id": "2026-09-30-openai-gpt-6-luna-max-codex",
      "releaseDate": "2026-09-22",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Codex",
          "url": "https://openai.com/codex/"
        },
        "id": "codex",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        },
        "version": "0.159.0"
      },
      "model": {
        "display": {
          "label": "GPT-6 Luna",
          "url": "https://developers.openai.com/api/docs/models/gpt-6-luna"
        },
        "id": "openai/gpt-6-luna",
        "organization": {
          "label": "OpenAI",
          "url": "https://openai.com"
        }
      },
      "metrics": {
        "accuracyPercent": 16.36,
        "accuracyCi95HalfWidthPercent": 2.72,
        "averageTrialDurationSeconds": 2284.1,
        "cachedInputTokens": 3778954219,
        "nTrials": 330,
        "outputTokens": 45490894,
        "passAt2": 0.2273,
        "passAt3": 0.2621,
        "passAt4": 0.2848,
        "passAt5": 0.303,
        "successes": 54,
        "totalCostUsd": 72.14,
        "totalTokens": 3909124893,
        "uncachedInputTokens": 3863633999
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/38bcca13-9ed9-5cf2-a864-d777ae4c76d1"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-30-openai-gpt-6-luna-max-codex.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-30-openai-gpt-6-luna-max-codex.json"
    },
    {
      "id": "2026-09-30-meta-muse-spark-1-3-xhigh-muse-code",
      "releaseDate": "2026-09-02",
      "reasoningEffort": "xhigh",
      "harness": {
        "display": {
          "label": "Muse Code",
          "url": "https://dev.meta.ai/docs/muse-code"
        },
        "id": "muse-code",
        "organization": {
          "label": "Meta",
          "url": "https://ai.meta.com"
        },
        "version": "1.4.1"
      },
      "model": {
        "display": {
          "label": "Muse Spark 1.3",
          "url": "https://ai.meta.com/tools/muse/"
        },
        "id": "meta/muse-spark-1.3",
        "organization": {
          "label": "Meta",
          "url": "https://ai.meta.com"
        }
      },
      "metrics": {
        "accuracyPercent": 14.55,
        "accuracyCi95HalfWidthPercent": 2.94,
        "averageTrialDurationSeconds": 4341.7,
        "cachedInputTokens": 8653482809,
        "nTrials": 330,
        "outputTokens": 87486990,
        "passAt2": 0.2197,
        "passAt3": 0.2652,
        "passAt4": 0.3,
        "passAt5": 0.3333,
        "successes": 48,
        "totalCostUsd": 0,
        "totalTokens": 9010923518,
        "uncachedInputTokens": 8923436528
      },
      "sourceFilterReasoningEffort": "xhigh",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/91784e40-2fb3-5fcf-afc0-74746868c015"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-30-meta-muse-spark-1-3-xhigh-muse-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-30-meta-muse-spark-1-3-xhigh-muse-code.json"
    },
    {
      "id": "2026-08-26-anthropic-claude-sonnet-5-max-claude-code",
      "releaseDate": "2026-06-30",
      "reasoningEffort": "max",
      "harness": {
        "display": {
          "label": "Claude Code",
          "url": "https://claude.com/product/claude-code"
        },
        "id": "claude-code",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        },
        "version": "2.1.231"
      },
      "model": {
        "display": {
          "label": "Sonnet 5",
          "url": "https://www.anthropic.com/news/claude-sonnet-5"
        },
        "id": "anthropic/claude-sonnet-5",
        "organization": {
          "label": "Anthropic",
          "url": "https://www.anthropic.com"
        }
      },
      "metrics": {
        "accuracyPercent": 12.42,
        "accuracyCi95HalfWidthPercent": 3.06,
        "averageTrialDurationSeconds": 6510.3,
        "cachedInputTokens": 21027162464,
        "nTrials": 330,
        "outputTokens": 116598866,
        "passAt2": 0.2045,
        "passAt3": 0.2621,
        "passAt4": 0.3091,
        "passAt5": 0.3485,
        "successes": 41,
        "totalCostUsd": 9603.86,
        "totalTokens": 21562180742,
        "uncachedInputTokens": 21445581876
      },
      "sourceFilterReasoningEffort": "max",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/a5758f1a-9ef2-4100-9893-34a99c99bd9c"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-anthropic-claude-sonnet-5-max-claude-code.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-anthropic-claude-sonnet-5-max-claude-code.json"
    },
    {
      "id": "2026-08-26-xai-grok-4-5-none-grok-build",
      "releaseDate": "2026-07-16",
      "reasoningEffort": "none",
      "harness": {
        "display": {
          "label": "Grok Build",
          "url": "https://x.ai/build"
        },
        "id": "grok-build",
        "organization": {
          "label": "xAI",
          "url": "https://x.ai"
        },
        "version": "1.0.5"
      },
      "model": {
        "display": {
          "label": "Grok 4.5",
          "url": "https://docs.x.ai/developers/models/grok-4.5"
        },
        "id": "xai/grok-4.5",
        "organization": {
          "label": "xAI",
          "url": "https://x.ai"
        }
      },
      "metrics": {
        "accuracyPercent": 12.42,
        "accuracyCi95HalfWidthPercent": 2.62,
        "averageTrialDurationSeconds": 3162.1,
        "cachedInputTokens": 3248613376,
        "nTrials": 330,
        "outputTokens": 32389328,
        "passAt2": 0.1833,
        "passAt3": 0.2227,
        "passAt4": 0.2515,
        "passAt5": 0.2727,
        "successes": 41,
        "totalCostUsd": 2094.11,
        "totalTokens": 3402536479,
        "uncachedInputTokens": 3370147151
      },
      "sourceFilterReasoningEffort": null,
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/51a17348-a8b8-4c5e-b8d2-73a9a724c3b6"
      ],
      "sourcePath": "leaderboard/submissions/2026-08-26-xai-grok-4-5-none-grok-build.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-08-26-xai-grok-4-5-none-grok-build.json"
    },
    {
      "id": "2026-09-02-gemini-gemini-3-7-flash-high-mini-swe-agent",
      "releaseDate": "2026-08-13",
      "reasoningEffort": "high",
      "harness": {
        "display": {
          "label": "mini-SWE-agent",
          "url": "https://github.com/SWE-agent/mini-swe-agent"
        },
        "id": "mini-swe-agent",
        "organization": {
          "label": "SWE-agent",
          "url": "https://github.com/SWE-agent"
        },
        "version": "2.4.6"
      },
      "model": {
        "display": {
          "label": "Gemini 3.7 Flash",
          "url": "https://ai.google.dev/gemini-api/docs/models/gemini-3.7-flash"
        },
        "id": "gemini/gemini-3.7-flash",
        "organization": {
          "label": "Google",
          "url": "https://google.com"
        }
      },
      "metrics": {
        "accuracyPercent": 11.21,
        "accuracyCi95HalfWidthPercent": 2.45,
        "averageTrialDurationSeconds": 1688.9,
        "cachedInputTokens": 10736925351,
        "nTrials": 330,
        "outputTokens": 52134001,
        "passAt2": 0.1636,
        "passAt3": 0.2,
        "passAt4": 0.2303,
        "passAt5": 0.2576,
        "successes": 37,
        "totalCostUsd": 1261.87,
        "totalTokens": 11137189802,
        "uncachedInputTokens": 11085055801
      },
      "sourceFilterReasoningEffort": "high",
      "sourceJobUrls": [
        "https://hub.harborframework.com/jobs/98bfa993-cbb6-5d79-a209-efdf3ed037a5"
      ],
      "sourcePath": "leaderboard/submissions/2026-09-02-gemini-gemini-3-7-flash-high-mini-swe-agent.json",
      "sourceUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench/3e9bf0ee42f7419722428680966435a7a71d0d40/leaderboard/submissions/2026-09-02-gemini-gemini-3-7-flash-high-mini-swe-agent.json"
    }
  ]
}
