{"source":"AIMap","schema":"benchmark/v1","updated":"2026-10-22","benchmark":{"id":"terminal-bench","name":"Terminal-Bench 2.0","category":"agent-coding","unit":"%","higherIsBetter":true,"status":"active","sources":["epoch"],"primaryVariant":null,"url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","year":2025,"desc":{"en":"Hard, realistic tasks an agent must finish in a sandboxed terminal (Stanford × Laude Institute).","zh-hans":"agent 需在沙箱终端里完成的高难度真实任务（斯坦福 × Laude Institute）。"},"measures":{"en":"Task accuracy for an agent + model pair. Data compiled by Epoch AI from tbench.ai.","zh-hans":"agent 与模型组合的任务正确率。数据由 Epoch AI 从 tbench.ai 汇编。"},"limits":{"en":"Agent harness matters as much as the model; the same model appears several times with different agents.","zh-hans":"agent 框架和模型同样重要，同一模型会搭配不同 agent 多次出现。"},"contamination":"low","contaminationNote":{"en":"Tasks carry a canary string.","zh-hans":"任务带有防训练标记。"},"stats":{"results":202,"models":45,"latest":"2026-05-14","bySourceType":{"compiled":5,"official":197}}},"sources":[{"id":"epoch","name":"Epoch AI Benchmarking Hub","license":"CC BY 4.0","policy":"open","attribution":"Epoch AI, 'Capabilities & benchmarking', https://epoch.ai/benchmarks (CC BY 4.0)","site":"https://epoch.ai/benchmarks"}],"results":[{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"compiled","setting":"Terminal-Bench 2.0; agent: CodeBrain-1","score":70.3,"unit":"%","rank":1,"rankComputedBySite":true,"date":"2026-02-10","sourceType":"compiled","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"compiled","setting":"Terminal-Bench 2.0; agent: CodeBrain-1","score":62.2,"unit":"%","rank":2,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"compiled","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-6","modelName":"Claude Sonnet 4.6","provider":"Anthropic","rawName":"Claude Sonnet 4.6","variant":"compiled","setting":"Terminal-Bench 2.0; agent: Simplai Agent; effort: no thinking","score":53.4,"unit":"%","rank":3,"rankComputedBySite":true,"date":"2026-05-14","sourceType":"compiled","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"compiled","setting":"Terminal-Bench 2.0; agent: MAYA; effort: unknown thinking","score":42.7,"unit":"%","rank":4,"rankComputedBySite":true,"date":"2026-01-04","sourceType":"compiled","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"compiled","setting":"Terminal-Bench 2.0; agent: spoox-m; effort: unknown thinking","score":34.8,"unit":"%","rank":5,"rankComputedBySite":true,"date":"2025-12-24","sourceType":"compiled","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-5","modelName":"GPT-5.5","provider":"OpenAI","rawName":"GPT-5.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: NexAU-AHE; effort: unknown thinking","score":84.72,"unit":"%","rank":1,"rankComputedBySite":true,"date":"2026-04-23","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-5","modelName":"GPT-5.5","provider":"OpenAI","rawName":"GPT-5.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Capy; effort: unknown thinking","score":83.15,"unit":"%","rank":2,"rankComputedBySite":true,"date":"2026-04-23","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-5","modelName":"GPT-5.5","provider":"OpenAI","rawName":"GPT-5.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI; effort: unknown thinking","score":82.25,"unit":"%","rank":3,"rankComputedBySite":true,"date":"2026-04-23","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-5","modelName":"GPT-5.5","provider":"OpenAI","rawName":"GPT-5.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex; effort: unknown thinking","score":82,"unit":"%","rank":4,"rankComputedBySite":true,"date":"2026-04-23","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-4","modelName":"GPT-5.4","provider":"OpenAI","rawName":"GPT-5.4","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: ForgeCode; effort: unknown thinking","score":81.8,"unit":"%","rank":5,"rankComputedBySite":true,"date":"2026-03-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-1-pro","modelName":"Gemini 3.1 Pro","provider":"Google","rawName":"Gemini 3.1 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: TongAgents","score":80.22,"unit":"%","rank":6,"rankComputedBySite":true,"date":"2026-02-19","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-7","modelName":"Claude Opus 4.7","provider":"Anthropic","rawName":"Claude Opus 4.7","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: WOZCODE; effort: unknown","score":80.22,"unit":"%","rank":6,"rankComputedBySite":true,"date":"2026-04-16","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: ForgeCode; effort: unknown thinking","score":79.8,"unit":"%","rank":8,"rankComputedBySite":true,"date":"2026-03-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: SageAgent","score":78.43,"unit":"%","rank":9,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-1-pro","modelName":"Gemini 3.1 Pro","provider":"Google","rawName":"Gemini 3.1 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Forge Code","score":78.4,"unit":"%","rank":10,"rankComputedBySite":true,"date":"2026-03-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-1-pro","modelName":"Gemini 3.1 Pro","provider":"Google","rawName":"Gemini 3.1 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: ForgeCode","score":78.4,"unit":"%","rank":10,"rankComputedBySite":true,"date":"2026-03-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Droid","score":77.3,"unit":"%","rank":12,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Meta-Harness; effort: unknown thinking","score":76.4,"unit":"%","rank":13,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: CodeBrain-1.5","score":75.84,"unit":"%","rank":14,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codelia","score":75.73,"unit":"%","rank":15,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Capy; effort: unknown thinking","score":75.28,"unit":"%","rank":16,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Simple Codex","score":75.06,"unit":"%","rank":17,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-1-pro","modelName":"Gemini 3.1 Pro","provider":"Google","rawName":"Gemini 3.1 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus-KIRA","score":74.83,"unit":"%","rank":18,"rankComputedBySite":true,"date":"2026-02-19","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus-KIRA; effort: unknown thinking","score":74.72,"unit":"%","rank":19,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mux","score":74.61,"unit":"%","rank":20,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: MAYA-V2; effort: unknown thinking","score":72.08,"unit":"%","rank":21,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: TongAgents; effort: unknown thinking","score":71.91,"unit":"%","rank":22,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: spoox-o-m","score":71.46,"unit":"%","rank":23,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Droid; effort: no thinking","score":69.9,"unit":"%","rank":24,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Droid; effort: unknown thinking","score":69.89,"unit":"%","rank":25,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Ante","score":69.43,"unit":"%","rank":26,"rankComputedBySite":true,"date":"2025-11-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: IndusAGI Coding Agent","score":69.14,"unit":"%","rank":27,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Crux; effort: unknown thinking","score":66.87,"unit":"%","rank":28,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mux; effort: unknown thinking","score":66.52,"unit":"%","rank":29,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-2-codex","modelName":"GPT-5.2 Codex","provider":"OpenAI","rawName":"GPT-5.2 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Deep Agents","score":66.52,"unit":"%","rank":29,"rankComputedBySite":true,"date":"2026-01-14","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-5","modelName":"GPT-5.5","provider":"OpenAI","rawName":"GPT-5.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: clnkr; effort: unknown thinking","score":66.07,"unit":"%","rank":31,"rankComputedBySite":true,"date":"2026-04-23","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: SageAgent","score":65.17,"unit":"%","rank":32,"rankComputedBySite":true,"date":"2025-11-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Droid; effort: unknown thinking","score":64.94,"unit":"%","rank":33,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Droid; effort: medium","score":64.9,"unit":"%","rank":34,"rankComputedBySite":true,"date":"2025-12-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-3-codex","modelName":"GPT-5.3 Codex","provider":"OpenAI","rawName":"GPT-5.3 Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":64.72,"unit":"%","rank":35,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-flash","modelName":"Gemini 3 Flash","provider":"Google","rawName":"Gemini 3 Flash","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Junie CLI","score":64.27,"unit":"%","rank":36,"rankComputedBySite":true,"date":"2025-12-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Droid; effort: unknown thinking","score":63.15,"unit":"%","rank":37,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Droid; effort: no thinking","score":63.1,"unit":"%","rank":38,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI; effort: unknown thinking","score":62.92,"unit":"%","rank":39,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":62.92,"unit":"%","rank":39,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI; effort: medium","score":62.9,"unit":"%","rank":41,"rankComputedBySite":true,"date":"2025-12-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: CodeBrain-1.5","score":62.25,"unit":"%","rank":42,"rankComputedBySite":true,"date":"2025-11-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: II-Agent","score":61.8,"unit":"%","rank":43,"rankComputedBySite":true,"date":"2025-11-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex-mini","modelName":"GPT-5.1-Codex-mini","provider":"OpenAI","rawName":"GPT-5.1-Codex-mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: hookele","score":61.57,"unit":"%","rank":44,"rankComputedBySite":true,"date":"2025-11-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-1-pro","modelName":"Gemini 3.1 Pro","provider":"Google","rawName":"Gemini 3.1 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Gemini CLI","score":61.42,"unit":"%","rank":45,"rankComputedBySite":true,"date":"2026-02-19","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Droid","score":61.12,"unit":"%","rank":46,"rankComputedBySite":true,"date":"2025-11-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mux; effort: unknown thinking","score":60.67,"unit":"%","rank":47,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex-max","modelName":"GPT-5.1-Codex-Max","provider":"OpenAI","rawName":"GPT-5.1-Codex-Max","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI","score":60.45,"unit":"%","rank":48,"rankComputedBySite":true,"date":"2025-11-19","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Letta Code; effort: unknown thinking","score":59.1,"unit":"%","rank":49,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Letta Code; effort: 128k thinking","score":59.1,"unit":"%","rank":49,"rankComputedBySite":true,"date":"2025-12-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mux; effort: unknown thinking","score":58.43,"unit":"%","rank":51,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Claude Code; effort: unknown thinking","score":57.98,"unit":"%","rank":52,"rankComputedBySite":true,"date":"2026-02-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex","modelName":"GPT-5.1-Codex","provider":"OpenAI","rawName":"GPT-5.1-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI","score":57.8,"unit":"%","rank":53,"rankComputedBySite":true,"date":"2025-11-16","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex","modelName":"GPT-5.1-Codex","provider":"OpenAI","rawName":"GPT-5.1-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Crux","score":57.75,"unit":"%","rank":54,"rankComputedBySite":true,"date":"2025-11-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":57.75,"unit":"%","rank":54,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-4-20","modelName":"Grok 4.20","provider":"xAI","rawName":"Grok 4.20","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Grok CLI","score":57.3,"unit":"%","rank":56,"rankComputedBySite":true,"date":"2026-03-10","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-4-20","modelName":"Grok 4.20","provider":"xAI","rawName":"Grok 4.20","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: grok-cli","score":57.3,"unit":"%","rank":56,"rankComputedBySite":true,"date":"2026-04-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":56.85,"unit":"%","rank":58,"rankComputedBySite":true,"date":"2025-11-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Letta Code","score":55.96,"unit":"%","rank":59,"rankComputedBySite":true,"date":"2025-11-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Goose; effort: unknown thinking","score":54.33,"unit":"%","rank":60,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":54.04,"unit":"%","rank":61,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex","modelName":"GPT-5.1-Codex","provider":"OpenAI","rawName":"GPT-5.1-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Letta Code","score":53.48,"unit":"%","rank":62,"rankComputedBySite":true,"date":"2025-11-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-6","modelName":"Claude Sonnet 4.6","provider":"Anthropic","rawName":"Claude Sonnet 4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Simplai Agent; effort: unknown thinking","score":53.43,"unit":"%","rank":63,"rankComputedBySite":true,"date":"2026-02-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"glm-5","modelName":"GLM-5","provider":"Zhipu AI (Z.ai)","rawName":"GLM-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":52.42,"unit":"%","rank":64,"rankComputedBySite":true,"date":"2026-02-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Claude Code; effort: unknown thinking","score":52.13,"unit":"%","rank":65,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking","score":51.91,"unit":"%","rank":66,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-flash","modelName":"Gemini 3 Flash","provider":"Google","rawName":"Gemini 3 Flash","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":51.69,"unit":"%","rank":67,"rankComputedBySite":true,"date":"2025-12-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenCode; effort: unknown thinking","score":51.69,"unit":"%","rank":67,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-flash","modelName":"Gemini 3 Flash","provider":"Google","rawName":"Gemini 3 Flash","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Gemini CLI     ","score":51,"unit":"%","rank":69,"rankComputedBySite":true,"date":"2025-12-23","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI; effort: unknown thinking","score":49.61,"unit":"%","rank":70,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Codex CLI      ; effort: medium","score":49.6,"unit":"%","rank":71,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1","modelName":"GPT-5.1","provider":"OpenAI","rawName":"GPT-5.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: medium","score":47.6,"unit":"%","rank":72,"rankComputedBySite":true,"date":"2025-11-16","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1","modelName":"GPT-5.1","provider":"OpenAI","rawName":"GPT-5.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ; effort: medium","score":47.6,"unit":"%","rank":72,"rankComputedBySite":true,"date":"2025-11-16","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1","modelName":"GPT-5.1","provider":"OpenAI","rawName":"GPT-5.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":47.58,"unit":"%","rank":74,"rankComputedBySite":true,"date":"2025-11-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-3-flash","modelName":"Gemini 3 Flash","provider":"Google","rawName":"Gemini 3 Flash","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Gemini CLI","score":47.42,"unit":"%","rank":75,"rankComputedBySite":true,"date":"2025-12-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: CAMEL-AI; effort: unknown thinking","score":46.52,"unit":"%","rank":76,"rankComputedBySite":true,"date":"2025-09-29","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"minimax-m2-7","modelName":"MiniMax-M2.7","provider":"MiniMax","rawName":"MiniMax-M2.7","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: IndusAGI Coding Agent","score":45.06,"unit":"%","rank":77,"rankComputedBySite":true,"date":"2026-03-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-codex","modelName":"GPT-5-Codex","provider":"OpenAI","rawName":"GPT-5-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI","score":44.33,"unit":"%","rank":78,"rankComputedBySite":true,"date":"2025-09-15","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-codex","modelName":"GPT-5-Codex","provider":"OpenAI","rawName":"GPT-5-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Codex CLI      ","score":44.3,"unit":"%","rank":79,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking","score":43.82,"unit":"%","rank":80,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ; effort: medium","score":43.8,"unit":"%","rank":81,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-codex","modelName":"GPT-5-Codex","provider":"OpenAI","rawName":"GPT-5-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":43.43,"unit":"%","rank":82,"rankComputedBySite":true,"date":"2025-09-15","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-codex","modelName":"GPT-5-Codex","provider":"OpenAI","rawName":"GPT-5-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":43.4,"unit":"%","rank":83,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"kimi-k2-5","modelName":"Kimi K2.5","provider":"Moonshot AI","rawName":"Kimi K2.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":43.2,"unit":"%","rank":84,"rankComputedBySite":true,"date":"2026-01-27","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex-mini","modelName":"GPT-5.1-Codex-mini","provider":"OpenAI","rawName":"GPT-5.1-Codex-mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Crux","score":43.15,"unit":"%","rank":85,"rankComputedBySite":true,"date":"2025-11-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Goose; effort: unknown thinking","score":43.15,"unit":"%","rank":85,"rankComputedBySite":true,"date":"2025-09-29","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex-mini","modelName":"GPT-5.1-Codex-mini","provider":"OpenAI","rawName":"GPT-5.1-Codex-mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Codex CLI      ","score":43.1,"unit":"%","rank":87,"rankComputedBySite":true,"date":"2025-11-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"minimax-m2-7","modelName":"MiniMax-M2.7","provider":"MiniMax","rawName":"MiniMax-M2.7","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Harness Agent","score":42.92,"unit":"%","rank":88,"rankComputedBySite":true,"date":"2026-03-18","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":42.81,"unit":"%","rank":89,"rankComputedBySite":true,"date":"2025-09-29","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ; effort: no thinking","score":42.8,"unit":"%","rank":90,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"minimax-m2-5","modelName":"MiniMax-M2.5","provider":"MiniMax","rawName":"MiniMax-M2.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: cchuter","score":42.7,"unit":"%","rank":91,"rankComputedBySite":true,"date":"2026-02-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: MAYA-V2; effort: unknown thinking","score":42.7,"unit":"%","rank":91,"rankComputedBySite":true,"date":"2025-09-29","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ; effort: no thinking","score":42.6,"unit":"%","rank":93,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking","score":42.58,"unit":"%","rank":94,"rankComputedBySite":true,"date":"2025-09-29","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent; effort: unknown thinking","score":42.53,"unit":"%","rank":95,"rankComputedBySite":true,"date":"2025-09-29","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ; effort: no thinking","score":42.5,"unit":"%","rank":96,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"minimax-m2-5","modelName":"MiniMax-M2.5","provider":"MiniMax","rawName":"MiniMax-M2.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":42.25,"unit":"%","rank":97,"rankComputedBySite":true,"date":"2026-02-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-codex","modelName":"GPT-5-Codex","provider":"OpenAI","rawName":"GPT-5-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent","score":41.35,"unit":"%","rank":98,"rankComputedBySite":true,"date":"2025-09-15","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-codex","modelName":"GPT-5-Codex","provider":"OpenAI","rawName":"GPT-5-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":41.3,"unit":"%","rank":99,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Claude Code    ; effort: no thinking","score":40.1,"unit":"%","rank":100,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Claude Code; effort: unknown thinking","score":40.06,"unit":"%","rank":101,"rankComputedBySite":true,"date":"2025-09-29","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"deepseek-v3-2","modelName":"DeepSeek-V3.2","provider":"DeepSeek","rawName":"DeepSeek-V3.2","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: Thinking; Novita","score":39.55,"unit":"%","rank":102,"rankComputedBySite":true,"date":"2025-12-01","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":38,"unit":"%","rank":103,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":37.98,"unit":"%","rank":104,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex","modelName":"GPT-5.1-Codex","provider":"OpenAI","rawName":"GPT-5.1-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":36.9,"unit":"%","rank":105,"rankComputedBySite":true,"date":"2025-11-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ","score":36.9,"unit":"%","rank":105,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking","score":36.85,"unit":"%","rank":107,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-1-codex","modelName":"GPT-5.1-Codex","provider":"OpenAI","rawName":"GPT-5.1-Codex","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":36.85,"unit":"%","rank":107,"rankComputedBySite":true,"date":"2025-11-12","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"minimax-m2-1","modelName":"MiniMax-M2.1","provider":"MiniMax","rawName":"MiniMax-M2.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Crux","score":36.63,"unit":"%","rank":109,"rankComputedBySite":true,"date":"2025-12-23","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"kimi-k2","modelName":"Kimi K2 Thinking","provider":"Moonshot AI","rawName":"Kimi K2 Thinking","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":35.73,"unit":"%","rank":110,"rankComputedBySite":true,"date":"2025-11-06","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"kimi-k2","modelName":"Kimi K2 Thinking","provider":"Moonshot AI","rawName":"Kimi K2 Thinking","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":35.7,"unit":"%","rank":111,"rankComputedBySite":true,"date":"2025-11-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Goose; effort: unknown thinking","score":35.51,"unit":"%","rank":112,"rankComputedBySite":true,"date":"2025-10-15","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ; effort: medium","score":35.2,"unit":"%","rank":113,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":35.17,"unit":"%","rank":114,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":35.1,"unit":"%","rank":115,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent; effort: unknown thinking","score":35.06,"unit":"%","rank":116,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Claude Code; effort: unknown thinking","score":34.83,"unit":"%","rank":117,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: spoox-o-m; effort: unknown thinking","score":34.83,"unit":"%","rank":117,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Claude Code    ","score":34.8,"unit":"%","rank":119,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent; effort: unknown thinking","score":33.93,"unit":"%","rank":120,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ; effort: medium","score":33.9,"unit":"%","rank":121,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"glm-4-7","modelName":"GLM-4.7","provider":"Zhipu AI (Z.ai)","rawName":"GLM-4.7","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":33.39,"unit":"%","rank":122,"rankComputedBySite":true,"date":"2025-12-22","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"glm-4-7","modelName":"GLM-4.7","provider":"Zhipu AI (Z.ai)","rawName":"GLM-4.7","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Crux","score":33.26,"unit":"%","rank":123,"rankComputedBySite":true,"date":"2025-12-22","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":32.64,"unit":"%","rank":124,"rankComputedBySite":true,"date":"2025-06-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":32.6,"unit":"%","rank":125,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Codex CLI      ; effort: medium","score":31.9,"unit":"%","rank":126,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI; effort: unknown thinking","score":31.85,"unit":"%","rank":127,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"minimax-m2","modelName":"MiniMax-M2","provider":"MiniMax","rawName":"MiniMax-M2","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":30.04,"unit":"%","rank":128,"rankComputedBySite":true,"date":"2025-10-27","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent; effort: unknown thinking","score":29.83,"unit":"%","rank":129,"rankComputedBySite":true,"date":"2025-10-15","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":29.8,"unit":"%","rank":130,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"minimax-m2-1","modelName":"MiniMax-M2.1","provider":"MiniMax","rawName":"MiniMax-M2.1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":29.21,"unit":"%","rank":131,"rankComputedBySite":true,"date":"2025-12-23","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking","score":29.21,"unit":"%","rank":131,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ; effort: medium","score":29.2,"unit":"%","rank":133,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":28.31,"unit":"%","rank":134,"rankComputedBySite":true,"date":"2025-10-15","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":28.3,"unit":"%","rank":135,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"kimi-k2-jul-2025","modelName":"Kimi K2 (Jul 2025)","provider":"Moonshot AI","rawName":"Kimi K2 (Jul 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":27.8,"unit":"%","rank":136,"rankComputedBySite":true,"date":"2025-11-01","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"kimi-k2-jul-2025","modelName":"Kimi K2 (Jul 2025)","provider":"Moonshot AI","rawName":"Kimi K2 (Jul 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":27.79,"unit":"%","rank":137,"rankComputedBySite":true,"date":"2025-07-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Claude Code; effort: unknown thinking","score":27.53,"unit":"%","rank":138,"rankComputedBySite":true,"date":"2025-10-15","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Claude Code    ","score":27.5,"unit":"%","rank":139,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-4","modelName":"Grok 4","provider":"xAI","rawName":"Grok 4","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ","score":27.2,"unit":"%","rank":140,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-4","modelName":"Grok 4","provider":"xAI","rawName":"Grok 4","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands","score":27.19,"unit":"%","rank":141,"rankComputedBySite":true,"date":"2025-07-09","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"qwen3-coder-480b-a35b","modelName":"Qwen3-Coder-480B-A35B","provider":"Alibaba (Qwen)","rawName":"Qwen3-Coder-480B-A35B","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Dakou Agent","score":27.19,"unit":"%","rank":141,"rankComputedBySite":true,"date":"2025-07-22","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"kimi-k2-jul-2025","modelName":"Kimi K2 (Jul 2025)","provider":"Moonshot AI","rawName":"Kimi K2 (Jul 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands","score":26.74,"unit":"%","rank":143,"rankComputedBySite":true,"date":"2025-07-11","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"kimi-k2-jul-2025","modelName":"Kimi K2 (Jul 2025)","provider":"Moonshot AI","rawName":"Kimi K2 (Jul 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ","score":26.7,"unit":"%","rank":144,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":26.1,"unit":"%","rank":145,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent","score":26.07,"unit":"%","rank":146,"rankComputedBySite":true,"date":"2025-06-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-code-fast-1","modelName":"grok-code-fast-1","provider":"xAI","rawName":"grok-code-fast-1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent","score":25.84,"unit":"%","rank":147,"rankComputedBySite":true,"date":"2025-08-28","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-code-fast-1","modelName":"grok-code-fast-1","provider":"xAI","rawName":"grok-code-fast-1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":25.8,"unit":"%","rank":148,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-4","modelName":"Grok 4","provider":"xAI","rawName":"Grok 4","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":25.4,"unit":"%","rank":149,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"qwen3-coder-480b-a35b","modelName":"Qwen3-Coder-480B-A35B","provider":"Alibaba (Qwen)","rawName":"Qwen3-Coder-480B-A35B","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ","score":25.4,"unit":"%","rank":149,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-4","modelName":"Grok 4","provider":"xAI","rawName":"Grok 4","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent","score":25.39,"unit":"%","rank":151,"rankComputedBySite":true,"date":"2025-07-09","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"qwen3-coder-480b-a35b","modelName":"Qwen3-Coder-480B-A35B","provider":"Alibaba (Qwen)","rawName":"Qwen3-Coder-480B-A35B","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands","score":25.39,"unit":"%","rank":151,"rankComputedBySite":true,"date":"2025-07-22","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"qwen-3-6-35b-a3b","modelName":"Qwen 3.6 35B-A3B","provider":"Alibaba (Qwen)","rawName":"Qwen 3.6 35B-A3B","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: little-coder","score":24.63,"unit":"%","rank":153,"rankComputedBySite":true,"date":"2026-04-16","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"glm-4-6","modelName":"GLM-4.6","provider":"Zhipu AI (Z.ai)","rawName":"GLM-4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":24.53,"unit":"%","rank":154,"rankComputedBySite":true,"date":"2025-09-30","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"glm-4-6","modelName":"GLM-4.6","provider":"Zhipu AI (Z.ai)","rawName":"GLM-4.6","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":24.5,"unit":"%","rank":155,"rankComputedBySite":true,"date":"2025-11-01","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":24.04,"unit":"%","rank":156,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ; effort: medium","score":24,"unit":"%","rank":157,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"qwen3-coder-480b-a35b","modelName":"Qwen3-Coder-480B-A35B","provider":"Alibaba (Qwen)","rawName":"Qwen3-Coder-480B-A35B","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":23.9,"unit":"%","rank":158,"rankComputedBySite":true,"date":"2025-11-01","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"qwen3-coder-480b-a35b","modelName":"Qwen3-Coder-480B-A35B","provider":"Alibaba (Qwen)","rawName":"Qwen3-Coder-480B-A35B","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":23.9,"unit":"%","rank":158,"rankComputedBySite":true,"date":"2025-07-22","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-4","modelName":"Grok 4","provider":"xAI","rawName":"Grok 4","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":23.15,"unit":"%","rank":160,"rankComputedBySite":true,"date":"2025-07-09","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-4","modelName":"Grok 4","provider":"xAI","rawName":"Grok 4","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":23.1,"unit":"%","rank":161,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent; effort: unknown thinking","score":22.25,"unit":"%","rank":162,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ; effort: medium","score":22.2,"unit":"%","rank":163,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: spoox-o-m; effort: unknown thinking","score":21.8,"unit":"%","rank":164,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Gemini CLI     ","score":19.6,"unit":"%","rank":165,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Gemini CLI","score":19.55,"unit":"%","rank":166,"rankComputedBySite":true,"date":"2025-06-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-oss-120b","modelName":"gpt-oss-120b","provider":"OpenAI","rawName":"gpt-oss-120b","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":18.7,"unit":"%","rank":167,"rankComputedBySite":true,"date":"2025-11-01","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-oss-120b","modelName":"gpt-oss-120b","provider":"OpenAI","rawName":"gpt-oss-120b","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":18.69,"unit":"%","rank":168,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-flash-sep-2025","modelName":"Gemini 2.5 Flash (Sep 2025)","provider":"Google","rawName":"Gemini 2.5 Flash (Sep 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":17.1,"unit":"%","rank":169,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-flash-jun-2025","modelName":"Gemini 2.5 Flash (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Flash (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent","score":17.08,"unit":"%","rank":170,"rankComputedBySite":true,"date":"2025-06-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-flash-sep-2025","modelName":"Gemini 2.5 Flash (Sep 2025)","provider":"Google","rawName":"Gemini 2.5 Flash (Sep 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":16.9,"unit":"%","rank":171,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-flash-jun-2025","modelName":"Gemini 2.5 Flash (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Flash (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":16.85,"unit":"%","rank":172,"rankComputedBySite":true,"date":"2025-06-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands","score":16.4,"unit":"%","rank":173,"rankComputedBySite":true,"date":"2025-06-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-flash-jun-2025","modelName":"Gemini 2.5 Flash (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Flash (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands","score":16.4,"unit":"%","rank":173,"rankComputedBySite":true,"date":"2025-06-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ","score":16.4,"unit":"%","rank":173,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-flash-sep-2025","modelName":"Gemini 2.5 Flash (Sep 2025)","provider":"Google","rawName":"Gemini 2.5 Flash (Sep 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ","score":16.4,"unit":"%","rank":173,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-flash-jun-2025","modelName":"Gemini 2.5 Flash (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Flash (Jun 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Gemini CLI","score":15.45,"unit":"%","rank":177,"rankComputedBySite":true,"date":"2025-06-17","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gemini-2-5-flash-sep-2025","modelName":"Gemini 2.5 Flash (Sep 2025)","provider":"Google","rawName":"Gemini 2.5 Flash (Sep 2025)","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Gemini CLI     ","score":15.4,"unit":"%","rank":178,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-code-fast-1","modelName":"grok-code-fast-1","provider":"xAI","rawName":"grok-code-fast-1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":14.2,"unit":"%","rank":179,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-oss-120b","modelName":"gpt-oss-120b","provider":"OpenAI","rawName":"gpt-oss-120b","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":14.2,"unit":"%","rank":179,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-oss-120b","modelName":"gpt-oss-120b","provider":"OpenAI","rawName":"gpt-oss-120b","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent; effort: unknown thinking","score":14.16,"unit":"%","rank":181,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"grok-code-fast-1","modelName":"grok-code-fast-1","provider":"xAI","rawName":"grok-code-fast-1","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":14.16,"unit":"%","rank":181,"rankComputedBySite":true,"date":"2025-08-28","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking","score":13.93,"unit":"%","rank":183,"rankComputedBySite":true,"date":"2025-10-15","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ","score":13.9,"unit":"%","rank":184,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Codex CLI      ; effort: medium","score":11.5,"unit":"%","rank":185,"rankComputedBySite":true,"date":"2025-11-04","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Codex CLI; effort: unknown thinking","score":11.46,"unit":"%","rank":186,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  OpenHands      ; effort: medium","score":9.9,"unit":"%","rank":187,"rankComputedBySite":true,"date":"2025-11-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking","score":9.89,"unit":"%","rank":188,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"qwen3-5-9b","modelName":"Qwen3.5-9B","provider":"Alibaba (Qwen)","rawName":"Qwen3.5-9B","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: little-coder","score":9.21,"unit":"%","rank":189,"rankComputedBySite":true,"date":"2026-03-02","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ; effort: medium","score":7.9,"unit":"%","rank":190,"rankComputedBySite":true,"date":"2025-10-31","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking","score":7.87,"unit":"%","rank":191,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ; effort: medium","score":7,"unit":"%","rank":192,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent; effort: unknown thinking","score":6.97,"unit":"%","rank":193,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-oss-20b","modelName":"gpt-oss-20b","provider":"OpenAI","rawName":"gpt-oss-20b","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Mini-SWE-Agent ","score":3.4,"unit":"%","rank":194,"rankComputedBySite":true,"date":"2025-11-03","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-oss-20b","modelName":"gpt-oss-20b","provider":"OpenAI","rawName":"gpt-oss-20b","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Mini-SWE-Agent","score":3.37,"unit":"%","rank":195,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-oss-20b","modelName":"gpt-oss-20b","provider":"OpenAI","rawName":"gpt-oss-20b","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent:  Terminus 2     ","score":3.1,"unit":"%","rank":196,"rankComputedBySite":true,"date":"2025-11-01","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"},{"model":"gpt-oss-20b","modelName":"gpt-oss-20b","provider":"OpenAI","rawName":"gpt-oss-20b","variant":"leaderboard","setting":"Terminal-Bench 2.0; agent: Terminus 2","score":3.07,"unit":"%","rank":197,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"official","source":"epoch","sourceUrl":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","collected":"2026-10-10"}]}