{
  "api_version": "1",
  "updated": "2026-10-18",
  "latest_event": "2026-10-17",
  "rights": "© Frontier Log, all rights reserved.",
  "benchmark": {
    "slug": "terminal-bench-1-0",
    "name": "Terminal-Bench 1.0",
    "family": "terminal-bench",
    "unit": "%",
    "higher_is_better": true,
    "status": "superseded",
    "benchmark_url": "https://www.tbench.ai",
    "introduced": "2025-05-19",
    "creator": "Stanford University and Laude Institute",
    "introduced_source": "https://www.tbench.ai/news/announcement",
    "settings": [
      {
        "setting": "default",
        "name": "",
        "url": "https://frontierlog.ai/benchmark/terminal-bench-1-0",
        "results": 5
      }
    ]
  },
  "total": 5,
  "results": [
    {
      "id": "claude-opus-4:terminal-bench-1-0:default:1",
      "model": "claude-opus-4",
      "model_name": "Claude Opus 4",
      "model_date": "2025-05-22",
      "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
      "benchmark": "terminal-bench-1-0",
      "benchmark_name": "Terminal-Bench 1.0",
      "setting": "default",
      "setting_name": "",
      "score": 43.2,
      "unit": "%",
      "printed": "43.2%",
      "reported_by": "anthropic",
      "label": "Claude Opus 4",
      "source": "https://www.anthropic.com/news/claude-4",
      "source_title": "Introducing Claude 4",
      "source_date": "2025-05-22",
      "where": "main benchmark table (HTML drawn by script), row 'Agentic terminal coding / Terminal-bench', column 'Claude Opus 4', first number of the cell",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "No extended thinking (appendix). Footnote 2: Claude Code as agent framework. Same number also in the intro sentence: 'leading on SWE-bench (72.5%) and Terminal-bench (43.2%)'. Printed as 'Terminal-bench' without version (2025).",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18"
    },
    {
      "id": "claude-opus-4:terminal-bench-1-0:default:2",
      "model": "claude-opus-4",
      "model_name": "Claude Opus 4",
      "model_date": "2025-05-22",
      "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
      "benchmark": "terminal-bench-1-0",
      "benchmark_name": "Terminal-Bench 1.0",
      "setting": "default",
      "setting_name": "",
      "score": 50.0,
      "unit": "%",
      "printed": "50.0%",
      "reported_by": "anthropic",
      "label": "Claude Opus 4",
      "source": "https://www.anthropic.com/news/claude-4",
      "source_title": "Introducing Claude 4",
      "source_date": "2025-05-22",
      "where": "main benchmark table, row 'Terminal-bench', column 'Claude Opus 4', second number of the cell (after the slash)",
      "own": true,
      "primary": false,
      "parallel": true,
      "notes": "Footnote 5: parallel test-time compute, sampling multiple sequences and selecting the best via an internal scoring model. Claude Code as agent framework.",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18"
    },
    {
      "id": "claude-opus-4:terminal-bench-1-0:default:3",
      "model": "claude-opus-4",
      "model_name": "Claude Opus 4",
      "model_date": "2025-05-22",
      "event": "https://frontierlog.ai/event/2025-05-22-claude-4",
      "benchmark": "terminal-bench-1-0",
      "benchmark_name": "Terminal-Bench 1.0",
      "setting": "default",
      "setting_name": "",
      "score": 39.2,
      "unit": "%",
      "printed": "39.2%",
      "reported_by": "anthropic",
      "label": "Opus 4",
      "source": "https://www.anthropic.com/news/claude-4",
      "source_title": "Introducing Claude 4",
      "source_date": "2025-05-22",
      "where": "methodology footnote 2 under the main table",
      "own": true,
      "primary": false,
      "parallel": false,
      "notes": "pass@1 with the same agent as non-Claude models (not Claude Code); no extended thinking.",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18"
    },
    {
      "id": "claude-opus-4-1:terminal-bench-1-0:default:1",
      "model": "claude-opus-4-1",
      "model_name": "Claude Opus 4.1",
      "model_date": "2025-08-05",
      "event": "https://frontierlog.ai/event/2025-08-05-claude-opus-4-1",
      "benchmark": "terminal-bench-1-0",
      "benchmark_name": "Terminal-Bench 1.0",
      "setting": "default",
      "setting_name": "",
      "score": 43.3,
      "unit": "%",
      "printed": "43.3%",
      "reported_by": "anthropic",
      "label": "Claude Opus 4.1",
      "source": "https://www.anthropic.com/news/claude-opus-4-1",
      "source_title": "Claude Opus 4.1",
      "source_date": "2025-08-05",
      "where": "benchmark table image, row 'Agentic terminal coding / Terminal-Bench', column 'Claude Opus 4.1'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "No extended thinking (appendix). Footnote 2: default agent framework ('Terminus 1'), averaged over 5 trials. Printed as 'Terminal-Bench' without version (2025).",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18",
      "source_image": "https://frontierlog.ai/sources/cebed70e8721d155.png",
      "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/bde326699c667506c87f74b09a6355961d29eb26-2600x2084.png"
    },
    {
      "id": "deepseek-v3-2-exp:terminal-bench-1-0:default:1",
      "model": "deepseek-v3-2-exp",
      "model_name": "DeepSeek-V3.2-Exp",
      "model_date": "2025-09-29",
      "event": "https://frontierlog.ai/event/2025-09-29-deepseek-v3-2-exp",
      "benchmark": "terminal-bench-1-0",
      "benchmark_name": "Terminal-Bench 1.0",
      "setting": "default",
      "setting_name": "",
      "score": 37.7,
      "unit": "%",
      "printed": "37.7",
      "reported_by": "deepseek",
      "label": "DeepSeek-V3.2-Exp",
      "source": "https://api-docs.deepseek.com/news/news250929/",
      "source_title": "Introducing DeepSeek-V3.2-Exp",
      "source_date": "2025-09-29",
      "where": "benchmark table image, section 'Code Agent', row 'Terminal-bench', column 'DeepSeek-V3.2-Exp'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "Row printed as 'Terminal-bench' with no version (2025). No metric or harness printed.",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18",
      "source_image": "https://frontierlog.ai/sources/4d9152a1f18ce8ee.webp",
      "source_image_original": "https://api-docs.deepseek.com/img/v3_2_benchmark.webp"
    }
  ]
}