{
  "api_version": "1",
  "updated": "2026-10-18",
  "latest_event": "2026-10-17",
  "rights": "© Frontier Log, all rights reserved.",
  "benchmark": {
    "slug": "terminal-bench-2-1",
    "name": "Terminal-Bench 2.1",
    "family": "terminal-bench",
    "unit": "%",
    "higher_is_better": true,
    "status": "superseded",
    "benchmark_url": "https://www.tbench.ai",
    "introduced": "2026-05-06",
    "creator": "Stanford University and Laude Institute",
    "introduced_source": "https://www.tbench.ai/news/terminal-bench-2-1",
    "settings": [
      {
        "setting": "default",
        "name": "",
        "url": "https://frontierlog.ai/benchmark/terminal-bench-2-1",
        "results": 4
      }
    ]
  },
  "total": 4,
  "results": [
    {
      "id": "claude-opus-4-8:terminal-bench-2-1:default:1",
      "model": "claude-opus-4-8",
      "model_name": "Claude Opus 4.8",
      "model_date": "2026-05-28",
      "event": "https://frontierlog.ai/event/2026-05-28-claude-opus-4-8",
      "benchmark": "terminal-bench-2-1",
      "benchmark_name": "Terminal-Bench 2.1",
      "setting": "default",
      "setting_name": "",
      "score": 74.6,
      "unit": "%",
      "printed": "74.6%",
      "reported_by": "anthropic",
      "label": "Opus 4.8",
      "source": "https://www.anthropic.com/news/claude-opus-4-8",
      "source_title": "Introducing Claude Opus 4.8",
      "source_date": "2026-05-28",
      "where": "benchmark table image, row 'Agentic terminal coding / Terminal-Bench 2.1', column 'Opus 4.8'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "Footnote: all models run with the Terminus-2 public harness. Opus 4.8 defaults to high effort (post text); effort for the table not stated.",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18",
      "source_image": "https://frontierlog.ai/sources/7ad9b784f2579c8a.png",
      "source_image_original": "https://www-cdn.anthropic.com/images/4zrzovbb/website/a9007019094f217e98cb8261a2765d7646c01708-2600x1392.png"
    },
    {
      "id": "gpt-5-6-sol:terminal-bench-2-1:default:1",
      "model": "gpt-5-6-sol",
      "model_name": "GPT-5.6 Sol",
      "model_date": "2026-07-09",
      "event": "https://frontierlog.ai/event/2026-07-09-gpt-5-6",
      "benchmark": "terminal-bench-2-1",
      "benchmark_name": "Terminal-Bench 2.1",
      "setting": "default",
      "setting_name": "",
      "score": 88.8,
      "unit": "%",
      "printed": "88.8%",
      "reported_by": "openai",
      "label": "GPT‑5.6 Sol",
      "source": "https://openai.com/index/gpt-5-6",
      "source_title": "GPT-5.6: Frontier intelligence that scales with your ambition",
      "source_date": "2026-07-09",
      "where": "coding eval table (HTML), row 'Terminal-Bench 2.1', column 'GPT‑5.6 Sol'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "Table also has a 'GPT‑5.6 Sol Ultra' column (separate row recorded). The fetched page text was truncated, so footnotes without a % sign are not visible in the extract.",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18"
    },
    {
      "id": "qwen3-8-max:terminal-bench-2-1:default:1",
      "model": "qwen3-8-max",
      "model_name": "Qwen3.8-Max",
      "model_date": "2026-08-03",
      "event": "https://frontierlog.ai/event/2026-08-03-qwen3-8-max",
      "benchmark": "terminal-bench-2-1",
      "benchmark_name": "Terminal-Bench 2.1",
      "setting": "default",
      "setting_name": "",
      "score": 86.6,
      "unit": "%",
      "printed": "86.6",
      "reported_by": "alibaba",
      "label": "Qwen3.8-Max",
      "source": "https://qwen.ai/blog?id=qwen3.8",
      "source_title": "Qwen3.8-Max: A New Bar for Coding and Cowork",
      "source_date": "2026-08-03",
      "where": "'Full Benchmark Table' (first, language/agent HTML table), section 'Coding Agent', row 'Terminal Bench 2.1', column 'Qwen3.8-Max'",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "Footnote 2: Evaluated with Claude Code (avg@10), 5-hour timeout, max_tokens=131,072. Default reasoning_effort per post is xhigh. Post HTML via https://qwen.ai/api/v2/article/?language=en-US&path=qwen3.8&type=qwen_ai.",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18"
    },
    {
      "id": "deepseek-v4-1-flash:terminal-bench-2-1:default:1",
      "model": "deepseek-v4-1-flash",
      "model_name": "DeepSeek-V4.1-Flash",
      "model_date": "2026-09-10",
      "event": "https://frontierlog.ai/event/2026-09-10-deepseek-v4-1-flash",
      "benchmark": "terminal-bench-2-1",
      "benchmark_name": "Terminal-Bench 2.1",
      "setting": "default",
      "setting_name": "",
      "score": 90.6,
      "unit": "%",
      "printed": "90.6",
      "reported_by": "deepseek",
      "label": "DeepSeek V4.1-Flash",
      "source": "https://api-docs.deepseek.com/news/news260910/",
      "source_title": "DeepSeek-V4.1-Flash: Smarter, Faster, More Efficient",
      "source_date": "2026-09-10",
      "where": "benchmark table image (v4.1_260910_benchmark_table_en), row 'Terminal-Bench 2.1', column 'DeepSeek V4.1-Flash' (highlighted)",
      "own": true,
      "primary": true,
      "parallel": false,
      "notes": "No harness or effort printed for this row. The post's bar chart covers Terminal-Bench 3.0, not 2.1.",
      "caveat": "",
      "human_verified": true,
      "human_verified_on": "2026-10-18",
      "source_image": "https://frontierlog.ai/sources/19d3b83b1b9e5f5b.png",
      "source_image_original": "https://api-docs.deepseek.com/img/v4.1_260910_benchmark_table_en.png"
    }
  ]
}