Terminal-Bench 4.1
The official leaderboard hasn’t scored a model released after GPT-6.2 Astra (November 2026). Labs report their own numbers for one newer model: Gemini 4 Neon at 65.2%.
The Terminal-Bench 4.0 tasks with hardened verifiers: agents complete 66 hard tasks in containerized terminals across software, ML, science, security, and operations. Test scripts verify each result, scored as resolution rate.
It kept the 4.0 tasks and hardened the verifiers of 44 of them against reward hacking, then regraded the saved 4.0 runs. 162 trials changed score, and every model stayed within its earlier 95% interval.
- Made by
- Stanford University, Harbor, and Laude Institute
- Released
- November 2026
- Kind
- Terminal tasks
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal
- Items from
- Another benchmark
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- Terminal-Bench 4.0
- In views
- Agents
- Strongest model tested
- GPT-6.2 Astra #1 on the capability index
- Last in a launch post
- November 2026
- Last model added
- November 2026
- Scores refreshed
- November 18th, 2026
Copilot Fusion leads with 71.2% Copilot Fusion leads with 71.2% MiMo-V2.6 Pro leads with 27% in mini-SWE-agent
Best score per entry from the benchmark's own leaderboard, read on November 18th, 2026. Where a model ran on its own with several settings, we keep its best. Every run is in the next chart. Agents other companies built around a model get a pale shade of that model's lab color. Gray marks blends and entries not in our model list, such as fine-tunes and older models. None of these count toward leaders. Agents built around a model, such as Claude Code + Kimi K3.1 Pro, are their own entries. We don't count them as the model's score. This view shows only models we track, each at its best. This view keeps only runs in mini-SWE-agent, a harness no model’s maker built. A maker’s own agent, such as Claude Code with a Claude model, is left out.
Copilot Fusion leads with 71.2%, claims included
Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model. Striped bars are claims, and dotted bars are reruns by Epoch AI or another board.
- 71.2 Copilot Fusion
- 65.8 Claude Opus 5.5
- 65.8 Claude Fable 5.2
- 64.5
Gemini 4 Carbon
- 63.9 Claude Sonnet 5.5
- 62.1 GPT-6.2 Astra
- 61.5 GPT-6.2 Sol
- 60.9 GPT-6.1 Sol
- 60.3 GPT-6 Astra
- 59.7 Claude Code + Kimi K3.1 Pro
- 57.6 Claude Fable 5.1
- 56.4
Gemini 4 Hydrogen
- 55.5
Gemini 4 Argon
- 53.3 Claude Opus 5
- 53.3 GPT-6.1 Terra
- 50.3 GPT-6 Sol
- 47.0
Muse Wonder
- 44.9 Claude Fable 5
- 41.2 Claude Code + GLM-5.3
- 40.0 Claude Haiku 5.5
- 38.8 Grok 4.7
- 38.5 GPT-5.6 Sol
- 38.5 Claude Code + DeepSeek V4.1 Flash 1017
- 37.9 Claude Code + Tencent Hy4
- 37.0 Claude Code + GLM-5.3-Flash
- 30.6
Muse Ember
- 27.6 Claude Code + Mistral Large 4
- 27.0 Claude Code + Qwen3.8 Max (0902)
- 27.0 MiMo-V2.6 Pro
- 23.6 Claude Opus 4.8
- 23.0 Claude Code + Kimi K3
- 21.8
Ember-1
- 21.5 GPT-5.6 Terra
- 19.7 Grok 4.6
- 19.1 Claude Code + Tencent Hy4 preview
- 18.2
Gemini 3.8 Flash
- 17.6 GPT-5.6 Luna
- 16.1 MiMo-V2.6 Flash
- 15.2 GPT-6 Luna
- 14.8
Muse Spark 1.3
- 13.0
Inkling 2 Small
- 12.4 Grok 4.5
- 12.4
Gemini 3.7 Flash
- 12.1 Claude Sonnet 5
- 7.58
MiniMax-M3
- 6.67 Claude Code + DeepSeek V4 Pro 0813
- 6.36 Beam 1.0
- 2.73
Gemini 3.1 Pro
- 1.82 GPT-5.4 Mini
- 0.91
Inkling-Small
- 0.30
Nemotron 3.5 Super VL
- 0.0
Inkling
- 1 Copilot Fusion max router 71.2%
- 2 Claude Opus 5.5 max 65.8%
- 3 Claude Fable 5.2 max 65.8%
- 4
Gemini 4 Carbon high 64.5%
- 5 Claude Sonnet 5.5 max 63.9%
- 6 GPT-6.2 Astra max 62.1%
- 7 GPT-6.2 Sol max 61.5%
- 8 GPT-6.1 Sol max 60.9%
- 9 GPT-6 Astra max 60.3%
- 10 Claude Code + Kimi K3.1 Pro max 59.7%
- 11 Claude Fable 5.1 xhigh 57.6%
- 12
Gemini 4 Hydrogen high 56.4%
- 13
Gemini 4 Argon high 55.5%
- 14 Claude Opus 5 xhigh 53.3%
- 15 GPT-6.1 Terra max 53.3%
- 16 GPT-6 Sol max 50.3%
- 17
Muse Wonder xhigh 47%
- 18 Claude Fable 5 max 44.9%
- 19 Claude Code + GLM-5.3 max 41.2%
- 20 Claude Haiku 5.5 max 40%
- 21 Grok 4.7 xhigh 38.8%
- 22 GPT-5.6 Sol max 38.5%
- 23 Claude Code + DeepSeek V4.1 Flash 1017 max 38.5%
- 24 Claude Code + Tencent Hy4 max 37.9%
- 25 Claude Code + GLM-5.3-Flash max 37%
- 26
Muse Ember xhigh 30.6%
- 27 Claude Code + Mistral Large 4 max 27.6%
- 28 Claude Code + Qwen3.8 Max (0902) max 27%
- 29 MiMo-V2.6 Pro none 27%
- 30 Claude Opus 4.8 max 23.6%
- 31 Claude Code + Kimi K3 max 23%
- 32
Ember-1 none 21.8%
- 33 GPT-5.6 Terra max 21.5%
- 34 Grok 4.6 high 19.7%
- 35 Claude Code + Tencent Hy4 preview max 19.1%
- 36
Gemini 3.8 Flash high 18.2%
- 37 GPT-5.6 Luna max 17.6%
- 38 MiMo-V2.6 Flash none 16.1%
- 39 GPT-6 Luna max 15.2%
- 40
Muse Spark 1.3 xhigh 14.8%
- 41
Inkling 2 Small xhigh 13%
- 42 Grok 4.5 high 12.4%
- 43
Gemini 3.7 Flash high 12.4%
- 44 Claude Sonnet 5 max 12.1%
- 45
MiniMax-M3 none 7.58%
- 46 Claude Code + DeepSeek V4 Pro 0813 max 6.67%
- 47 Beam 1.0 none 6.36%
- 48
Gemini 3.1 Pro high 2.73%
- 49 GPT-5.4 Mini xhigh 1.82%
- 50
Inkling-Small none 0.91%
- 51
Nemotron 3.5 Super VL none 0.30%
- 52
Inkling none 0%
Source: Official Terminal-Bench 4.1 leaderboard
- 71.2 Copilot Fusion
- 65.8 Claude Opus 5.5
- 65.8 Claude Fable 5.2
- 64.5
Gemini 4 Carbon
- 63.9 Claude Sonnet 5.5
- 62.1 GPT-6.2 Astra
- 61.5 GPT-6.2 Sol
- 60.9 GPT-6.1 Sol
- 60.3 GPT-6 Astra
- 57.6 Claude Fable 5.1
- 56.4
Gemini 4 Hydrogen
- 55.5
Gemini 4 Argon
- 53.3 Claude Opus 5
- 53.3 GPT-6.1 Terra
- 50.3 GPT-6 Sol
- 47.0
Muse Wonder
- 44.9 Claude Fable 5
- 40.0 Claude Haiku 5.5
- 38.8 Grok 4.7
- 38.5 GPT-5.6 Sol
- 30.6
Muse Ember
- 27.0 MiMo-V2.6 Pro
- 23.6 Claude Opus 4.8
- 21.8
Ember-1
- 21.5 GPT-5.6 Terra
- 19.7 Grok 4.6
- 18.2
Gemini 3.8 Flash
- 17.6 GPT-5.6 Luna
- 16.1 MiMo-V2.6 Flash
- 15.2 GPT-6 Luna
- 14.8
Muse Spark 1.3
- 13.0
Inkling 2 Small
- 12.4 Grok 4.5
- 12.4
Gemini 3.7 Flash
- 12.1 Claude Sonnet 5
- 7.58
MiniMax-M3
- 6.36 Beam 1.0
- 2.73
Gemini 3.1 Pro
- 1.82 GPT-5.4 Mini
- 0.91
Inkling-Small
- 0.30
Nemotron 3.5 Super VL
- 0.0
Inkling
- 1 Copilot Fusion max router 71.2%
- 2 Claude Opus 5.5 max 65.8%
- 3 Claude Fable 5.2 max 65.8%
- 4
Gemini 4 Carbon high 64.5%
- 5 Claude Sonnet 5.5 max 63.9%
- 6 GPT-6.2 Astra max 62.1%
- 7 GPT-6.2 Sol max 61.5%
- 8 GPT-6.1 Sol max 60.9%
- 9 GPT-6 Astra max 60.3%
- 10 Claude Fable 5.1 xhigh 57.6%
- 11
Gemini 4 Hydrogen high 56.4%
- 12
Gemini 4 Argon high 55.5%
- 13 Claude Opus 5 xhigh 53.3%
- 14 GPT-6.1 Terra max 53.3%
- 15 GPT-6 Sol max 50.3%
- 16
Muse Wonder xhigh 47%
- 17 Claude Fable 5 max 44.9%
- 18 Claude Haiku 5.5 max 40%
- 19 Grok 4.7 xhigh 38.8%
- 20 GPT-5.6 Sol max 38.5%
- 21
Muse Ember xhigh 30.6%
- 22 MiMo-V2.6 Pro none 27%
- 23 Claude Opus 4.8 max 23.6%
- 24
Ember-1 none 21.8%
- 25 GPT-5.6 Terra max 21.5%
- 26 Grok 4.6 high 19.7%
- 27
Gemini 3.8 Flash high 18.2%
- 28 GPT-5.6 Luna max 17.6%
- 29 MiMo-V2.6 Flash none 16.1%
- 30 GPT-6 Luna max 15.2%
- 31
Muse Spark 1.3 xhigh 14.8%
- 32
Inkling 2 Small xhigh 13%
- 33 Grok 4.5 high 12.4%
- 34
Gemini 3.7 Flash high 12.4%
- 35 Claude Sonnet 5 max 12.1%
- 36
MiniMax-M3 none 7.58%
- 37 Beam 1.0 none 6.36%
- 38
Gemini 3.1 Pro high 2.73%
- 39 GPT-5.4 Mini xhigh 1.82%
- 40
Inkling-Small none 0.91%
- 41
Nemotron 3.5 Super VL none 0.30%
- 42
Inkling none 0%
Source: Official Terminal-Bench 4.1 leaderboard
- 1 MiMo-V2.6 Pro none 27%
- 2
Ember-1 none 21.8%
- 3
Gemini 3.8 Flash high 18.2%
- 4 MiMo-V2.6 Flash none 16.1%
- 5
Inkling 2 Small xhigh 13%
- 6
Gemini 3.7 Flash high 12.4%
- 7
MiniMax-M3 none 7.58%
- 8 Beam 1.0 none 6.36%
- 9
Inkling-Small none 0.91%
- 10
Nemotron 3.5 Super VL none 0.30%
- 11
Inkling none 0%
Source: Official Terminal-Bench 4.1 leaderboard
- 71.2 Copilot Fusion
- 65.8 Claude Opus 5.5
- 65.8 Claude Fable 5.2
- 65.2
Gemini 4 Neon
- 64.5
Gemini 4 Carbon
- 63.9 Claude Sonnet 5.5
- 62.1 GPT-6.2 Astra
- 61.5 GPT-6.2 Sol
- 60.9 GPT-6.1 Sol
- 60.3 GPT-6 Astra
- 57.6 Claude Fable 5.1
- 57.0
Gemini 4 Hydrogen
- 55.5
Gemini 4 Argon
- 53.3 Claude Opus 5
- 53.3 GPT-6.1 Terra
- 50.3 GPT-6 Sol
- 47.0
Muse Wonder
- 44.9 Claude Fable 5
- 40.0 Claude Haiku 5.5
- 38.8 Grok 4.7
- 38.5 GPT-5.6 Sol
- 30.6
Muse Ember
- 27.0 MiMo-V2.6 Pro
- 23.6 Claude Opus 4.8
- 21.8
Ember-1
- 21.5 GPT-5.6 Terra
- 19.7 Grok 4.6
- 18.2
Gemini 3.8 Flash
- 17.6 GPT-5.6 Luna
- 16.1 MiMo-V2.6 Flash
- 15.2 GPT-6 Luna
- 14.8
Muse Spark 1.3
- 13.0
Inkling 2 Small
- 12.4 Grok 4.5
- 12.4
Gemini 3.7 Flash
- 12.1 Claude Sonnet 5
- 7.58
MiniMax-M3
- 6.36 Beam 1.0
- 2.73
Gemini 3.1 Pro
- 1.82 GPT-5.4 Mini
- 0.91
Inkling-Small
- 0.30
Nemotron 3.5 Super VL
- 0.0
Inkling
- 1 Copilot Fusion max router 71.2%
- 2 Claude Opus 5.5 max 65.8%
- 3 Claude Fable 5.2 max 65.8%
- 4
Gemini 4 Neon 65.2%
- 5
Gemini 4 Carbon high 64.5%
- 6 Claude Sonnet 5.5 max 63.9%
- 7 GPT-6.2 Astra max 62.1%
- 8 GPT-6.2 Sol max 61.5%
- 9 GPT-6.1 Sol max 60.9%
- 10 GPT-6 Astra max 60.3%
- 11 Claude Fable 5.1 xhigh 57.6%
- 12
Gemini 4 Hydrogen 57%
- 13
Gemini 4 Argon high 55.5%
- 14 Claude Opus 5 xhigh 53.3%
- 15 GPT-6.1 Terra max 53.3%
- 16 GPT-6 Sol max 50.3%
- 17
Muse Wonder xhigh 47%
- 18 Claude Fable 5 max 44.9%
- 19 Claude Haiku 5.5 max 40%
- 20 Grok 4.7 xhigh 38.8%
- 21 GPT-5.6 Sol max 38.5%
- 22
Muse Ember xhigh 30.6%
- 23 MiMo-V2.6 Pro none 27%
- 24 Claude Opus 4.8 max 23.6%
- 25
Ember-1 none 21.8%
- 26 GPT-5.6 Terra max 21.5%
- 27 Grok 4.6 high 19.7%
- 28
Gemini 3.8 Flash high 18.2%
- 29 GPT-5.6 Luna max 17.6%
- 30 MiMo-V2.6 Flash none 16.1%
- 31 GPT-6 Luna max 15.2%
- 32
Muse Spark 1.3 xhigh 14.8%
- 33
Inkling 2 Small xhigh 13%
- 34 Grok 4.5 high 12.4%
- 35
Gemini 3.7 Flash high 12.4%
- 36 Claude Sonnet 5 max 12.1%
- 37
MiniMax-M3 none 7.58%
- 38 Beam 1.0 none 6.36%
- 39
Gemini 3.1 Pro high 2.73%
- 40 GPT-5.4 Mini xhigh 1.82%
- 41
Inkling-Small none 0.91%
- 42
Nemotron 3.5 Super VL none 0.30%
- 43
Inkling none 0%
Sources: Official Terminal-Bench 4.1 leaderboard and labs’ model cards
| # | Entry | Score ± 95% CI | Setting | Agent by | Released | Run on | Source |
|---|---|---|---|---|---|---|---|
| 1 | Copilot Fusion router | 71.2% ±2.9 | max, GitHub Copilot | GitHub | Oct 2026 | Nov 2026 | Official leaderboard |
| 2 | Claude Opus 5.5 | 65.8% ±3.1 | max, Claude Code | Anthropic | Sep 2026 | Sep 2026 | Official leaderboard |
| 3 | Claude Fable 5.2 | 65.8% ±2.9 | max, Claude Code | Anthropic | Nov 2026 | Nov 2026 | Official leaderboard |
| 4 | | 64.5% ±3.2 | high, Antigravity CLI | Oct 2026 | Oct 2026 | Official leaderboard | |
| 5 | Claude Sonnet 5.5 | 63.9% ±3 | max, Claude Code | Anthropic | Sep 2026 | Sep 2026 | Official leaderboard |
| 6 | GPT-6.2 Astra | 62.1% ±2.3 | max, Codex | OpenAI | Nov 2026 | Nov 2026 | Official leaderboard |
| 7 | GPT-6.2 Sol | 61.5% ±2.7 | max, Codex | OpenAI | Nov 2026 | Nov 2026 | Official leaderboard |
| 8 | GPT-6.1 Sol | 60.9% ±3 | max, Codex | OpenAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 9 | GPT-6 Astra | 60.3% ±2.8 | max, Codex | OpenAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 10 | Claude Code + Kimi K3.1 Pro | 59.7% ±4.2 | max | Anthropic | Oct 2026 | Oct 2026 | Official leaderboard |
| 11 | Claude Fable 5.1 | 57.6% ±3.3 | xhigh, Claude Code | Anthropic | Sep 2026 | Sep 2026 | Official leaderboard |
| 12 | | 56.4% ±3 | high, Antigravity CLI | Nov 2026 | Nov 2026 | Official leaderboard | |
| 13 | | 55.5% ±2.8 | high, Antigravity CLI | Sep 2026 | Sep 2026 | Official leaderboard | |
| 14 | Claude Opus 5 | 53.3% ±3.1 | xhigh, Claude Code | Anthropic | Jul 2026 | Jul 2026 | Official leaderboard |
| 15 | GPT-6.1 Terra | 53.3% ±3.1 | max, Codex | OpenAI | Oct 2026 | Oct 2026 | Official leaderboard |
| 16 | GPT-6 Sol | 50.3% ±3.2 | max, Codex | OpenAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 17 | | 47% ±3.8 | xhigh, Muse Code | Meta | Oct 2026 | Oct 2026 | Official leaderboard |
| 18 | Claude Fable 5 | 44.9% ±3.8 | max, Claude Code | Anthropic | Jun 2026 | Jun 2026 | Official leaderboard |
| 19 | Claude Code + GLM-5.3 | 41.2% ±3.3 | max | Anthropic | Aug 2026 | Aug 2026 | Official leaderboard |
| 20 | Claude Haiku 5.5 | 40% ±3.6 | max, Claude Code | Anthropic | Oct 2026 | Oct 2026 | Official leaderboard |
| 21 | Grok 4.7 | 38.8% ±3.6 | xhigh, Grok Build | xAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 22 | GPT-5.6 Sol | 38.5% ±3.8 | max, Codex | OpenAI | Jul 2026 | Jun 2026 | Official leaderboard |
| 23 | Claude Code + DeepSeek V4.1 Flash 1017 | 38.5% ±3.8 | max | Anthropic | Oct 2026 | Oct 2026 | Official leaderboard |
| 24 | Claude Code + Tencent Hy4 | 37.9% ±3.5 | max | Anthropic | Nov 2026 | Nov 2026 | Official leaderboard |
| 25 | Claude Code + GLM-5.3-Flash | 37% ±3.6 | max | Anthropic | Aug 2026 | Aug 2026 | Official leaderboard |
| 26 | | 30.6% ±3.8 | xhigh, Muse Code | Meta | Nov 2026 | Nov 2026 | Official leaderboard |
| 27 | Claude Code + Mistral Large 4 | 27.6% ±3.9 | max | Anthropic | Oct 2026 | Oct 2026 | Official leaderboard |
| 28 | Claude Code + Qwen3.8 Max (0902) | 27% ±3.9 | max | Anthropic | Sep 2026 | Sep 2026 | Official leaderboard |
| 29 | MiMo-V2.6 Pro | 27% ±3.6 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 30 | Claude Opus 4.8 | 23.6% ±3.6 | max, Claude Code | Anthropic | May 2026 | May 2026 | Official leaderboard |
| 31 | Claude Code + Kimi K3 | 23% ±3.7 | max | Anthropic | Jul 2026 | Jul 2026 | Official leaderboard |
| 32 | | 21.8% ±3.5 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 33 | GPT-5.6 Terra | 21.5% ±3.3 | max, Codex | OpenAI | Jul 2026 | Jun 2026 | Official leaderboard |
| 34 | Grok 4.6 | 19.7% ±3.1 | high, Grok Build | xAI | Aug 2026 | Aug 2026 | Official leaderboard |
| 35 | Claude Code + Tencent Hy4 preview | 19.1% ±3.1 | max | Anthropic | Aug 2026 | Aug 2026 | Official leaderboard |
| 36 | | 18.2% ±3.4 | high, mini-SWE-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 37 | GPT-5.6 Luna | 17.6% ±2.9 | max, Codex | OpenAI | Jul 2026 | Jun 2026 | Official leaderboard |
| 38 | MiMo-V2.6 Flash | 16.1% ±3.4 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 39 | GPT-6 Luna | 15.2% ±2.7 | max, Codex | OpenAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 40 | | 14.8% ±3 | xhigh, Muse Code | Meta | Sep 2026 | Sep 2026 | Official leaderboard |
| 41 | | 13% ±2.9 | xhigh, mini-swe-agent | SWE-agent | Nov 2026 | Nov 2026 | Official leaderboard |
| 42 | Grok 4.5 | 12.4% ±2.6 | high, Grok Build | xAI | Jul 2026 | Jul 2026 | Official leaderboard |
| 43 | | 12.4% ±2.5 | high, mini-SWE-agent | SWE-agent | Aug 2026 | Aug 2026 | Official leaderboard |
| 44 | Claude Sonnet 5 | 12.1% ±3 | max, Claude Code | Anthropic | Jun 2026 | Jun 2026 | Official leaderboard |
| 45 | | 7.58% ±2.8 | none, mini-swe-agent | SWE-agent | Jun 2026 | Jun 2026 | Official leaderboard |
| 46 | Claude Code + DeepSeek V4 Pro 0813 | 6.67% ±2.5 | max | Anthropic | Aug 2026 | Aug 2026 | Official leaderboard |
| 47 | Beam 1.0 | 6.36% ±1.7 | none, mini-swe-agent | SWE-agent | Oct 2026 | Oct 2026 | Official leaderboard |
| 48 | | 2.73% ±1.7 | high, Antigravity CLI | Feb 2026 | Feb 2026 | Official leaderboard | |
| 49 | GPT-5.4 Mini | 1.82% ±1.3 | xhigh, Codex | OpenAI | Mar 2026 | Mar 2026 | Official leaderboard |
| 50 | | 0.91% ±0.9 | none, mini-swe-agent | SWE-agent | Jul 2026 | Jul 2026 | Official leaderboard |
| 51 | | 0.30% ±0.6 | none, mini-swe-agent | SWE-agent | Oct 2026 | Oct 2026 | Official leaderboard |
| 52 | | 0% | none, mini-swe-agent | SWE-agent | Jul 2026 | Jul 2026 | Official leaderboard |
| # | Model | Score ± 95% CI | Setting | Agent by | Released | Run on | Source |
|---|---|---|---|---|---|---|---|
| 1 | Copilot Fusion router | 71.2% ±2.9 | max, GitHub Copilot | GitHub | Oct 2026 | Nov 2026 | Official leaderboard |
| 2 | Claude Opus 5.5 | 65.8% ±3.1 | max, Claude Code | Anthropic | Sep 2026 | Sep 2026 | Official leaderboard |
| 3 | Claude Fable 5.2 | 65.8% ±2.9 | max, Claude Code | Anthropic | Nov 2026 | Nov 2026 | Official leaderboard |
| 4 | | 64.5% ±3.2 | high, Antigravity CLI | Oct 2026 | Oct 2026 | Official leaderboard | |
| 5 | Claude Sonnet 5.5 | 63.9% ±3 | max, Claude Code | Anthropic | Sep 2026 | Sep 2026 | Official leaderboard |
| 6 | GPT-6.2 Astra | 62.1% ±2.3 | max, Codex | OpenAI | Nov 2026 | Nov 2026 | Official leaderboard |
| 7 | GPT-6.2 Sol | 61.5% ±2.7 | max, Codex | OpenAI | Nov 2026 | Nov 2026 | Official leaderboard |
| 8 | GPT-6.1 Sol | 60.9% ±3 | max, Codex | OpenAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 9 | GPT-6 Astra | 60.3% ±2.8 | max, Codex | OpenAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 10 | Claude Fable 5.1 | 57.6% ±3.3 | xhigh, Claude Code | Anthropic | Sep 2026 | Sep 2026 | Official leaderboard |
| 11 | | 56.4% ±3 | high, Antigravity CLI | Nov 2026 | Nov 2026 | Official leaderboard | |
| 12 | | 55.5% ±2.8 | high, Antigravity CLI | Sep 2026 | Sep 2026 | Official leaderboard | |
| 13 | Claude Opus 5 | 53.3% ±3.1 | xhigh, Claude Code | Anthropic | Jul 2026 | Jul 2026 | Official leaderboard |
| 14 | GPT-6.1 Terra | 53.3% ±3.1 | max, Codex | OpenAI | Oct 2026 | Oct 2026 | Official leaderboard |
| 15 | GPT-6 Sol | 50.3% ±3.2 | max, Codex | OpenAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 16 | | 47% ±3.8 | xhigh, Muse Code | Meta | Oct 2026 | Oct 2026 | Official leaderboard |
| 17 | Claude Fable 5 | 44.9% ±3.8 | max, Claude Code | Anthropic | Jun 2026 | Jun 2026 | Official leaderboard |
| 18 | Claude Haiku 5.5 | 40% ±3.6 | max, Claude Code | Anthropic | Oct 2026 | Oct 2026 | Official leaderboard |
| 19 | Grok 4.7 | 38.8% ±3.6 | xhigh, Grok Build | xAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 20 | GPT-5.6 Sol | 38.5% ±3.8 | max, Codex | OpenAI | Jul 2026 | Jun 2026 | Official leaderboard |
| 21 | | 30.6% ±3.8 | xhigh, Muse Code | Meta | Nov 2026 | Nov 2026 | Official leaderboard |
| 22 | MiMo-V2.6 Pro | 27% ±3.6 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 23 | Claude Opus 4.8 | 23.6% ±3.6 | max, Claude Code | Anthropic | May 2026 | May 2026 | Official leaderboard |
| 24 | | 21.8% ±3.5 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 25 | GPT-5.6 Terra | 21.5% ±3.3 | max, Codex | OpenAI | Jul 2026 | Jun 2026 | Official leaderboard |
| 26 | Grok 4.6 | 19.7% ±3.1 | high, Grok Build | xAI | Aug 2026 | Aug 2026 | Official leaderboard |
| 27 | | 18.2% ±3.4 | high, mini-SWE-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 28 | GPT-5.6 Luna | 17.6% ±2.9 | max, Codex | OpenAI | Jul 2026 | Jun 2026 | Official leaderboard |
| 29 | MiMo-V2.6 Flash | 16.1% ±3.4 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 30 | GPT-6 Luna | 15.2% ±2.7 | max, Codex | OpenAI | Sep 2026 | Sep 2026 | Official leaderboard |
| 31 | | 14.8% ±3 | xhigh, Muse Code | Meta | Sep 2026 | Sep 2026 | Official leaderboard |
| 32 | | 13% ±2.9 | xhigh, mini-swe-agent | SWE-agent | Nov 2026 | Nov 2026 | Official leaderboard |
| 33 | Grok 4.5 | 12.4% ±2.6 | high, Grok Build | xAI | Jul 2026 | Jul 2026 | Official leaderboard |
| 34 | | 12.4% ±2.5 | high, mini-SWE-agent | SWE-agent | Aug 2026 | Aug 2026 | Official leaderboard |
| 35 | Claude Sonnet 5 | 12.1% ±3 | max, Claude Code | Anthropic | Jun 2026 | Jun 2026 | Official leaderboard |
| 36 | | 7.58% ±2.8 | none, mini-swe-agent | SWE-agent | Jun 2026 | Jun 2026 | Official leaderboard |
| 37 | Beam 1.0 | 6.36% ±1.7 | none, mini-swe-agent | SWE-agent | Oct 2026 | Oct 2026 | Official leaderboard |
| 38 | | 2.73% ±1.7 | high, Antigravity CLI | Feb 2026 | Feb 2026 | Official leaderboard | |
| 39 | GPT-5.4 Mini | 1.82% ±1.3 | xhigh, Codex | OpenAI | Mar 2026 | Mar 2026 | Official leaderboard |
| 40 | | 0.91% ±0.9 | none, mini-swe-agent | SWE-agent | Jul 2026 | Jul 2026 | Official leaderboard |
| 41 | | 0.30% ±0.6 | none, mini-swe-agent | SWE-agent | Oct 2026 | Oct 2026 | Official leaderboard |
| 42 | | 0% | none, mini-swe-agent | SWE-agent | Jul 2026 | Jul 2026 | Official leaderboard |
| # | Model | Score ± 95% CI | Setting | Agent by | Released | Run on | Source |
|---|---|---|---|---|---|---|---|
| 1 | MiMo-V2.6 Pro | 27% ±3.6 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 2 | | 21.8% ±3.5 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 3 | | 18.2% ±3.4 | high, mini-SWE-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 4 | MiMo-V2.6 Flash | 16.1% ±3.4 | none, mini-swe-agent | SWE-agent | Sep 2026 | Sep 2026 | Official leaderboard |
| 5 | | 13% ±2.9 | xhigh, mini-swe-agent | SWE-agent | Nov 2026 | Nov 2026 | Official leaderboard |
| 6 | | 12.4% ±2.5 | high, mini-SWE-agent | SWE-agent | Aug 2026 | Aug 2026 | Official leaderboard |
| 7 | | 7.58% ±2.8 | none, mini-swe-agent | SWE-agent | Jun 2026 | Jun 2026 | Official leaderboard |
| 8 | Beam 1.0 | 6.36% ±1.7 | none, mini-swe-agent | SWE-agent | Oct 2026 | Oct 2026 | Official leaderboard |
| 9 | | 0.91% ±0.9 | none, mini-swe-agent | SWE-agent | Jul 2026 | Jul 2026 | Official leaderboard |
| 10 | | 0.30% ±0.6 | none, mini-swe-agent | SWE-agent | Oct 2026 | Oct 2026 | Official leaderboard |
| 11 | | 0% | none, mini-swe-agent | SWE-agent | Jul 2026 | Jul 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Copilot Fusion router | 71.2% | max, GitHub Copilot | Oct 2026 | Official leaderboard |
| 2 | Claude Opus 5.5 | 65.8% | max, Claude Code | Sep 2026 | Official leaderboard |
| 3 | Claude Fable 5.2 | 65.8% | max, Claude Code | Nov 2026 | Official leaderboard |
| 4 | 65.2% | Nov 2026 | Lab model card | ||
| 5 | 64.5% | high, Antigravity CLI | Oct 2026 | Official leaderboard | |
| 6 | Claude Sonnet 5.5 | 63.9% | max, Claude Code | Sep 2026 | Official leaderboard |
| 7 | GPT-6.2 Astra | 62.1% | max, Codex | Nov 2026 | Official leaderboard |
| 8 | GPT-6.2 Sol | 61.5% | max, Codex | Nov 2026 | Official leaderboard |
| 9 | GPT-6.1 Sol | 60.9% | max, Codex | Sep 2026 | Official leaderboard |
| 10 | GPT-6 Astra | 60.3% | max, Codex | Sep 2026 | Official leaderboard |
| 11 | Claude Fable 5.1 | 57.6% | xhigh, Claude Code | Sep 2026 | Official leaderboard |
| 12 | 57% | Nov 2026 | Lab model card | ||
| 13 | 55.5% | high, Antigravity CLI | Sep 2026 | Official leaderboard | |
| 14 | Claude Opus 5 | 53.3% | xhigh, Claude Code | Jul 2026 | Official leaderboard |
| 15 | GPT-6.1 Terra | 53.3% | max, Codex | Oct 2026 | Official leaderboard |
| 16 | GPT-6 Sol | 50.3% | max, Codex | Sep 2026 | Official leaderboard |
| 17 | 47% | xhigh, Muse Code | Oct 2026 | Official leaderboard | |
| 18 | Claude Fable 5 | 44.9% | max, Claude Code | Jun 2026 | Official leaderboard |
| 19 | Claude Haiku 5.5 | 40% | max, Claude Code | Oct 2026 | Official leaderboard |
| 20 | Grok 4.7 | 38.8% | xhigh, Grok Build | Sep 2026 | Official leaderboard |
| 21 | GPT-5.6 Sol | 38.5% | max, Codex | Jul 2026 | Official leaderboard |
| 22 | 30.6% | xhigh, Muse Code | Nov 2026 | Official leaderboard | |
| 23 | MiMo-V2.6 Pro | 27% | none, mini-swe-agent | Sep 2026 | Official leaderboard |
| 24 | Claude Opus 4.8 | 23.6% | max, Claude Code | May 2026 | Official leaderboard |
| 25 | 21.8% | none, mini-swe-agent | Sep 2026 | Official leaderboard | |
| 26 | GPT-5.6 Terra | 21.5% | max, Codex | Jul 2026 | Official leaderboard |
| 27 | Grok 4.6 | 19.7% | high, Grok Build | Aug 2026 | Official leaderboard |
| 28 | 18.2% | high, mini-SWE-agent | Sep 2026 | Official leaderboard | |
| 29 | GPT-5.6 Luna | 17.6% | max, Codex | Jul 2026 | Official leaderboard |
| 30 | MiMo-V2.6 Flash | 16.1% | none, mini-swe-agent | Sep 2026 | Official leaderboard |
| 31 | GPT-6 Luna | 15.2% | max, Codex | Sep 2026 | Official leaderboard |
| 32 | 14.8% | xhigh, Muse Code | Sep 2026 | Official leaderboard | |
| 33 | 13% | xhigh, mini-swe-agent | Nov 2026 | Official leaderboard | |
| 34 | Grok 4.5 | 12.4% | high, Grok Build | Jul 2026 | Official leaderboard |
| 35 | 12.4% | high, mini-SWE-agent | Aug 2026 | Official leaderboard | |
| 36 | Claude Sonnet 5 | 12.1% | max, Claude Code | Jun 2026 | Official leaderboard |
| 37 | 7.58% | none, mini-swe-agent | Jun 2026 | Official leaderboard | |
| 38 | Beam 1.0 | 6.36% | none, mini-swe-agent | Oct 2026 | Official leaderboard |
| 39 | 2.73% | high, Antigravity CLI | Feb 2026 | Official leaderboard | |
| 40 | GPT-5.4 Mini | 1.82% | xhigh, Codex | Mar 2026 | Official leaderboard |
| 41 | 0.91% | none, mini-swe-agent | Jul 2026 | Official leaderboard | |
| 42 | 0.30% | none, mini-swe-agent | Oct 2026 | Official leaderboard | |
| 43 | 0% | none, mini-swe-agent | Jul 2026 | Official leaderboard |
Labs report scores for 2 models in their cards. Switch the scores above to rank them together.
Pareto: accuracy against time
Each point is one agent, model and effort. The board plots total run time; this plots the mean time per trial, the same frontier at a fixed trial count. The line joins the runs that no other run beats on both measures. We redrew this chart from the numbers on the official leaderboard.
| Model | Setting | Accuracy | Mean time per trial |
|---|---|---|---|
| Copilot Fusion router | max, GitHub Copilot | 71.2% | 4,234 s |
| Claude Opus 5.5 | max, Claude Code | 65.8% | 4,790 s |
| Claude Fable 5.2 | max, Claude Code | 65.8% | 7,333 s |
| high, Antigravity CLI | 64.6% | 3,496 s | |
| Claude Sonnet 5.5 | max, Claude Code | 63.9% | 7,432 s |
| GPT-6.2 Astra | max, Codex | 62.1% | 1,350 s |
| GPT-6.2 Sol | max, Codex | 61.5% | 1,214 s |
| GPT-6.1 Sol | max, Codex | 60.9% | 3,177 s |
| GPT-6 Astra | max, Codex | 60.3% | 2,796 s |
| GPT-6 Astra | high, Codex | 60% | 2,114 s |
| Claude Code + Kimi K3.1 Pro | max, Claude Code | 59.7% | 123.2 s |
| GPT-6 Astra | xhigh, Codex | 59.7% | 2,209 s |
| Claude Fable 5.1 | xhigh, Claude Code | 57.6% | 3,202 s |
| Claude Fable 5.1 | max, Claude Code | 57.3% | 3,894 s |
| high, Antigravity CLI | 56.4% | 1,926 s | |
| high, Antigravity CLI | 55.5% | 125.2 s | |
| GPT-6 Astra | medium, Codex | 55.2% | 1,886 s |
| Claude Fable 5.1 | high, Claude Code | 54.9% | 3,018 s |
| Claude Fable 5.1 | medium, Claude Code | 53.9% | 2,528 s |
| GPT-6.1 Terra | max, Codex | 53.3% | 123.1 s |
| Claude Opus 5 | xhigh, Claude Code | 53.3% | 4,515 s |
| Claude Opus 5 | max, Claude Code | 52.1% | 4,793 s |
| GPT-6 Astra | low, Codex | 50.9% | 1,684 s |
| GPT-6 Sol | max, Codex | 50.3% | 2,662 s |
| Claude Opus 5 | high, Claude Code | 49.7% | 3,822 s |
| xhigh, Muse Code | 47% | 2,828 s | |
| Claude Opus 5 | medium, Claude Code | 45.2% | 3,293 s |
| Claude Fable 5 | max, Claude Code | 44.9% | 4,203 s |
| Claude Fable 5.1 | low, Claude Code | 43.3% | 2,291 s |
| Claude Code + GLM-5.3 | max, Claude Code | 41.2% | 5,831 s |
| Claude Haiku 5.5 | max, Claude Code | 40% | 140.5 s |
| Grok 4.7 | xhigh, Grok Build | 38.8% | 5,701 s |
| Claude Code + DeepSeek V4.1 Flash 1017 | max, Claude Code | 38.5% | 148.4 s |
| GPT-5.6 Sol | max, Codex | 38.5% | 2,388 s |
| Claude Code + Tencent Hy4 | max, Claude Code | 37.9% | 126.2 s |
| Claude Code + GLM-5.3-Flash | max, Claude Code | 37% | 120.4 s |
| Claude Code + GLM-5.3-Flash | none, Claude Code | 35.5% | 6,210 s |
| Claude Opus 5 | low, Claude Code | 34.9% | 2,947 s |
| xhigh, Muse Code | 30.6% | 2,288 s | |
| Claude Code + Mistral Large 4 | max, Claude Code | 27.6% | 142.7 s |
| MiMo-V2.6 Pro | none, mini-swe-agent | 27% | 168.2 s |
| Claude Code + Qwen3.8 Max (0902) | max, Claude Code | 27% | 9,953 s |
| Claude Opus 4.8 | max, Claude Code | 23.6% | 5,163 s |
| Claude Code + Kimi K3 | max, Claude Code | 23% | 141.9 s |
| none, mini-swe-agent | 21.8% | 4,960 s | |
| GPT-5.6 Terra | max, Codex | 21.5% | 2,515 s |
| Grok 4.6 | high, Grok Build | 19.7% | 2,352 s |
| Claude Code + Tencent Hy4 preview | max, Claude Code | 19.1% | 6,967 s |
| high, mini-SWE-agent | 18.2% | 1,995 s | |
| GPT-5.6 Luna | max, Codex | 17.6% | 4,088 s |
| MiMo-V2.6 Flash | none, mini-swe-agent | 16.1% | 160.8 s |
| GPT-6 Luna | max, Codex | 15.2% | 2,284 s |
| xhigh, Muse Code | 14.9% | 4,342 s | |
| xhigh, mini-swe-agent | 13% | 7,273 s | |
| high, mini-SWE-agent | 12.4% | 1,689 s | |
| Grok 4.5 | high, Grok Build | 12.4% | 3,162 s |
| Claude Sonnet 5 | max, Claude Code | 12.1% | 6,510 s |
| none, mini-swe-agent | 7.6% | 204 s | |
| Claude Code + DeepSeek V4 Pro 0813 | max, Claude Code | 6.7% | 138.2 s |
| Beam 1.0 | none, mini-swe-agent | 6.4% | 178.6 s |
| high, Antigravity CLI | 2.7% | 4,583 s | |
| GPT-5.4 Mini | xhigh, Codex | 1.8% | 141.5 s |
| none, mini-swe-agent | 0.9% | 141.3 s | |
| none, mini-swe-agent | 0.3% | 135.8 s | |
| none, mini-swe-agent | 0% | 174.1 s |
Pareto: accuracy against time
Each point is one agent, model and effort. The board plots total run time; this plots the mean time per trial, the same frontier at a fixed trial count. The line joins the runs that no other run beats on both measures. We redrew this chart from the numbers on the official leaderboard.
| Model | Setting | Accuracy | Mean time per trial |
|---|---|---|---|
| Copilot Fusion router | max, GitHub Copilot | 71.2% | 4,234 s |
| Claude Opus 5.5 | max, Claude Code | 65.8% | 4,790 s |
| Claude Fable 5.2 | max, Claude Code | 65.8% | 7,333 s |
| high, Antigravity CLI | 64.6% | 3,496 s | |
| Claude Sonnet 5.5 | max, Claude Code | 63.9% | 7,432 s |
| GPT-6.2 Astra | max, Codex | 62.1% | 1,350 s |
| GPT-6.2 Sol | max, Codex | 61.5% | 1,214 s |
| GPT-6.1 Sol | max, Codex | 60.9% | 3,177 s |
| GPT-6 Astra | max, Codex | 60.3% | 2,796 s |
| GPT-6 Astra | high, Codex | 60% | 2,114 s |
| GPT-6 Astra | xhigh, Codex | 59.7% | 2,209 s |
| Claude Fable 5.1 | xhigh, Claude Code | 57.6% | 3,202 s |
| Claude Fable 5.1 | max, Claude Code | 57.3% | 3,894 s |
| high, Antigravity CLI | 56.4% | 1,926 s | |
| high, Antigravity CLI | 55.5% | 125.2 s | |
| GPT-6 Astra | medium, Codex | 55.2% | 1,886 s |
| Claude Fable 5.1 | high, Claude Code | 54.9% | 3,018 s |
| Claude Fable 5.1 | medium, Claude Code | 53.9% | 2,528 s |
| GPT-6.1 Terra | max, Codex | 53.3% | 123.1 s |
| Claude Opus 5 | xhigh, Claude Code | 53.3% | 4,515 s |
| Claude Opus 5 | max, Claude Code | 52.1% | 4,793 s |
| GPT-6 Astra | low, Codex | 50.9% | 1,684 s |
| GPT-6 Sol | max, Codex | 50.3% | 2,662 s |
| Claude Opus 5 | high, Claude Code | 49.7% | 3,822 s |
| xhigh, Muse Code | 47% | 2,828 s | |
| Claude Opus 5 | medium, Claude Code | 45.2% | 3,293 s |
| Claude Fable 5 | max, Claude Code | 44.9% | 4,203 s |
| Claude Fable 5.1 | low, Claude Code | 43.3% | 2,291 s |
| Claude Haiku 5.5 | max, Claude Code | 40% | 140.5 s |
| Grok 4.7 | xhigh, Grok Build | 38.8% | 5,701 s |
| GPT-5.6 Sol | max, Codex | 38.5% | 2,388 s |
| Claude Opus 5 | low, Claude Code | 34.9% | 2,947 s |
| xhigh, Muse Code | 30.6% | 2,288 s | |
| MiMo-V2.6 Pro | none, mini-swe-agent | 27% | 168.2 s |
| Claude Opus 4.8 | max, Claude Code | 23.6% | 5,163 s |
| none, mini-swe-agent | 21.8% | 4,960 s | |
| GPT-5.6 Terra | max, Codex | 21.5% | 2,515 s |
| Grok 4.6 | high, Grok Build | 19.7% | 2,352 s |
| high, mini-SWE-agent | 18.2% | 1,995 s | |
| GPT-5.6 Luna | max, Codex | 17.6% | 4,088 s |
| MiMo-V2.6 Flash | none, mini-swe-agent | 16.1% | 160.8 s |
| GPT-6 Luna | max, Codex | 15.2% | 2,284 s |
| xhigh, Muse Code | 14.9% | 4,342 s | |
| xhigh, mini-swe-agent | 13% | 7,273 s | |
| high, mini-SWE-agent | 12.4% | 1,689 s | |
| Grok 4.5 | high, Grok Build | 12.4% | 3,162 s |
| Claude Sonnet 5 | max, Claude Code | 12.1% | 6,510 s |
| none, mini-swe-agent | 7.6% | 204 s | |
| Beam 1.0 | none, mini-swe-agent | 6.4% | 178.6 s |
| high, Antigravity CLI | 2.7% | 4,583 s | |
| GPT-5.4 Mini | xhigh, Codex | 1.8% | 141.5 s |
| none, mini-swe-agent | 0.9% | 141.3 s | |
| none, mini-swe-agent | 0.3% | 135.8 s | |
| none, mini-swe-agent | 0% | 174.1 s |
Pareto: accuracy against time
Each point is one agent, model and effort. The board plots total run time; this plots the mean time per trial, the same frontier at a fixed trial count. The line joins the runs that no other run beats on both measures. We redrew this chart from the numbers on the official leaderboard.
| Model | Setting | Accuracy | Mean time per trial |
|---|---|---|---|
| MiMo-V2.6 Pro | none, mini-swe-agent | 27% | 168.2 s |
| none, mini-swe-agent | 21.8% | 4,960 s | |
| high, mini-SWE-agent | 18.2% | 1,995 s | |
| MiMo-V2.6 Flash | none, mini-swe-agent | 16.1% | 160.8 s |
| xhigh, mini-swe-agent | 13% | 7,273 s | |
| high, mini-SWE-agent | 12.4% | 1,689 s | |
| none, mini-swe-agent | 7.6% | 204 s | |
| Beam 1.0 | none, mini-swe-agent | 6.4% | 178.6 s |
| none, mini-swe-agent | 0.9% | 141.3 s | |
| none, mini-swe-agent | 0.3% | 135.8 s | |
| none, mini-swe-agent | 0% | 174.1 s |
The best score, 71.2%, costs $6.02 a task
Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Terminal-Bench 4.1 leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Copilot Fusion router | max | GitHub Copilot | 71.2% | $6.02 | 243,149 |
| Claude Opus 5.5 | max | Claude Code | 65.8% | $14 | 269,135 |
| Claude Fable 5.2 | max | Claude Code | 65.8% | $50 | 491,352 |
| high | Antigravity CLI | 64.5% | $4.12 | 166,082 | |
| Claude Sonnet 5.5 | max | Claude Code | 63.9% | $22 | 565,388 |
| GPT-6.2 Astra | max | Codex | 62.1% | $4.27 | 49,894 |
| GPT-6.2 Sol | max | Codex | 61.5% | $1.06 | 48,561 |
| GPT-6.1 Sol | max | Codex | 60.9% | $1.92 | 94,892 |
| GPT-6 Astra | max | Codex | 60.3% | $9.90 | 72,694 |
| GPT-6 Astra | high | Codex | 60% | $6.88 | 41,527 |
| GPT-6 Astra | xhigh | Codex | 59.7% | $7.12 | 45,304 |
| Claude Code + Kimi K3.1 Pro | max | Claude Code | 59.7% | $3.29 | 80,956 |
| Claude Fable 5.1 | xhigh | Claude Code | 57.6% | $15 | 152,809 |
| Claude Fable 5.1 | max | Claude Code | 57.3% | $19 | 191,277 |
| high | Antigravity CLI | 56.4% | $2.38 | 119,482 | |
| high | Antigravity CLI | 55.5% | $3.36 | 162,988 | |
| GPT-6 Astra | medium | Codex | 55.1% | $5.80 | 32,638 |
| Claude Fable 5.1 | high | Claude Code | 54.9% | $12 | 113,421 |
| Claude Fable 5.1 | medium | Claude Code | 53.9% | $8.59 | 78,763 |
| Claude Opus 5 | xhigh | Claude Code | 53.3% | $18 | 179,281 |
| GPT-6.1 Terra | max | Codex | 53.3% | $2.65 | 100,138 |
| Claude Opus 5 | max | Claude Code | 52.1% | $18 | 200,037 |
| GPT-6 Astra | low | Codex | 50.9% | $4.72 | 24,207 |
| GPT-6 Sol | max | Codex | 50.3% | $4.40 | 92,642 |
| Claude Opus 5 | high | Claude Code | 49.7% | $14 | 144,242 |
| xhigh | Muse Code | 47% | $7.43 | 129,324 | |
| Claude Opus 5 | medium | Claude Code | 45.1% | $9.67 | 101,450 |
| Claude Fable 5 | max | Claude Code | 44.9% | $22 | 177,636 |
| Claude Fable 5.1 | low | Claude Code | 43.3% | $7.15 | 60,409 |
| Claude Code + GLM-5.3 | max | Claude Code | 41.2% | $8.27 | 208,061 |
| Claude Haiku 5.5 | max | Claude Code | 40% | $2.89 | 386,483 |
| Grok 4.7 | xhigh | Grok Build | 38.8% | $11 | 69,627 |
| GPT-5.6 Sol | max | Codex | 38.5% | $7.70 | 71,065 |
| Claude Code + DeepSeek V4.1 Flash 1017 | max | Claude Code | 38.5% | $0.56 | 497,265 |
| Claude Code + Tencent Hy4 | max | Claude Code | 37.9% | $1.05 | 368,811 |
| Claude Code + GLM-5.3-Flash | max | Claude Code | 37% | $1.14 | 459,702 |
| Claude Code + GLM-5.3-Flash | none | Claude Code | 35.5% | $0.36 | 135,014 |
| Claude Opus 5 | low | Claude Code | 34.9% | $7.25 | 72,229 |
| xhigh | Muse Code | 30.6% | $0.48 | 165,351 | |
| Claude Code + Mistral Large 4 | max | Claude Code | 27.6% | $2.18 | 253,650 |
| MiMo-V2.6 Pro | none | mini-swe-agent | 27% | $0.86 | 245,849 |
| Claude Code + Qwen3.8 Max (0902) | max | Claude Code | 27% | $14 | 252,708 |
| Claude Opus 4.8 | max | Claude Code | 23.6% | $20 | 269,131 |
| Claude Code + Kimi K3 | max | Claude Code | 23% | $5.82 | 195,427 |
| none | mini-swe-agent | 21.8% | $14 | 187,695 | |
| GPT-5.6 Terra | max | Codex | 21.5% | $5.25 | 100,681 |
| Grok 4.6 | high | Grok Build | 19.7% | $11 | 98,843 |
| Claude Code + Tencent Hy4 preview | max | Claude Code | 19.1% | $1.99 | 309,296 |
| high | mini-SWE-agent | 18.2% | $5.54 | 206,232 | |
| GPT-5.6 Luna | max | Codex | 17.6% | $1.05 | 185,705 |
| MiMo-V2.6 Flash | none | mini-swe-agent | 16.1% | $0.59 | 200,090 |
| GPT-6 Luna | max | Codex | 15.2% | $0.22 | 137,851 |
| xhigh | Muse Code | 14.8% | 265,112 | ||
| xhigh | mini-swe-agent | 13% | $2.92 | 309,788 | |
| Grok 4.5 | high | Grok Build | 12.4% | $6.35 | 98,149 |
| high | mini-SWE-agent | 12.4% | $3.82 | 157,982 | |
| Claude Sonnet 5 | max | Claude Code | 12.1% | $29 | 353,330 |
| none | mini-swe-agent | 7.58% | $3.48 | 234,421 | |
| Claude Code + DeepSeek V4 Pro 0813 | max | Claude Code | 6.67% | $2.20 | 245,205 |
| Beam 1.0 | none | mini-swe-agent | 6.36% | $0.67 | 64,523 |
| high | Antigravity CLI | 2.73% | $11 | 87,678 | |
| GPT-5.4 Mini | xhigh | Codex | 1.82% | $2.04 | 169,597 |
| none | mini-swe-agent | 0.91% | $0.15 | 20,104 | |
| none | mini-swe-agent | 0.30% | $0.51 | 159,709 | |
| none | mini-swe-agent | 0% | $3.06 | 39,308 |
The best score, 71.2%, costs $6.02 a task
Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Terminal-Bench 4.1 leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Copilot Fusion router | max | GitHub Copilot | 71.2% | $6.02 | 243,149 |
| Claude Opus 5.5 | max | Claude Code | 65.8% | $14 | 269,135 |
| Claude Fable 5.2 | max | Claude Code | 65.8% | $50 | 491,352 |
| high | Antigravity CLI | 64.5% | $4.12 | 166,082 | |
| Claude Sonnet 5.5 | max | Claude Code | 63.9% | $22 | 565,388 |
| GPT-6.2 Astra | max | Codex | 62.1% | $4.27 | 49,894 |
| GPT-6.2 Sol | max | Codex | 61.5% | $1.06 | 48,561 |
| GPT-6.1 Sol | max | Codex | 60.9% | $1.92 | 94,892 |
| GPT-6 Astra | max | Codex | 60.3% | $9.90 | 72,694 |
| GPT-6 Astra | high | Codex | 60% | $6.88 | 41,527 |
| GPT-6 Astra | xhigh | Codex | 59.7% | $7.12 | 45,304 |
| Claude Fable 5.1 | xhigh | Claude Code | 57.6% | $15 | 152,809 |
| Claude Fable 5.1 | max | Claude Code | 57.3% | $19 | 191,277 |
| high | Antigravity CLI | 56.4% | $2.38 | 119,482 | |
| high | Antigravity CLI | 55.5% | $3.36 | 162,988 | |
| GPT-6 Astra | medium | Codex | 55.1% | $5.80 | 32,638 |
| Claude Fable 5.1 | high | Claude Code | 54.9% | $12 | 113,421 |
| Claude Fable 5.1 | medium | Claude Code | 53.9% | $8.59 | 78,763 |
| Claude Opus 5 | xhigh | Claude Code | 53.3% | $18 | 179,281 |
| GPT-6.1 Terra | max | Codex | 53.3% | $2.65 | 100,138 |
| Claude Opus 5 | max | Claude Code | 52.1% | $18 | 200,037 |
| GPT-6 Astra | low | Codex | 50.9% | $4.72 | 24,207 |
| GPT-6 Sol | max | Codex | 50.3% | $4.40 | 92,642 |
| Claude Opus 5 | high | Claude Code | 49.7% | $14 | 144,242 |
| xhigh | Muse Code | 47% | $7.43 | 129,324 | |
| Claude Opus 5 | medium | Claude Code | 45.1% | $9.67 | 101,450 |
| Claude Fable 5 | max | Claude Code | 44.9% | $22 | 177,636 |
| Claude Fable 5.1 | low | Claude Code | 43.3% | $7.15 | 60,409 |
| Claude Haiku 5.5 | max | Claude Code | 40% | $2.89 | 386,483 |
| Grok 4.7 | xhigh | Grok Build | 38.8% | $11 | 69,627 |
| GPT-5.6 Sol | max | Codex | 38.5% | $7.70 | 71,065 |
| Claude Opus 5 | low | Claude Code | 34.9% | $7.25 | 72,229 |
| xhigh | Muse Code | 30.6% | $0.48 | 165,351 | |
| MiMo-V2.6 Pro | none | mini-swe-agent | 27% | $0.86 | 245,849 |
| Claude Opus 4.8 | max | Claude Code | 23.6% | $20 | 269,131 |
| none | mini-swe-agent | 21.8% | $14 | 187,695 | |
| GPT-5.6 Terra | max | Codex | 21.5% | $5.25 | 100,681 |
| Grok 4.6 | high | Grok Build | 19.7% | $11 | 98,843 |
| high | mini-SWE-agent | 18.2% | $5.54 | 206,232 | |
| GPT-5.6 Luna | max | Codex | 17.6% | $1.05 | 185,705 |
| MiMo-V2.6 Flash | none | mini-swe-agent | 16.1% | $0.59 | 200,090 |
| GPT-6 Luna | max | Codex | 15.2% | $0.22 | 137,851 |
| xhigh | Muse Code | 14.8% | 265,112 | ||
| xhigh | mini-swe-agent | 13% | $2.92 | 309,788 | |
| Grok 4.5 | high | Grok Build | 12.4% | $6.35 | 98,149 |
| high | mini-SWE-agent | 12.4% | $3.82 | 157,982 | |
| Claude Sonnet 5 | max | Claude Code | 12.1% | $29 | 353,330 |
| none | mini-swe-agent | 7.58% | $3.48 | 234,421 | |
| Beam 1.0 | none | mini-swe-agent | 6.36% | $0.67 | 64,523 |
| high | Antigravity CLI | 2.73% | $11 | 87,678 | |
| GPT-5.4 Mini | xhigh | Codex | 1.82% | $2.04 | 169,597 |
| none | mini-swe-agent | 0.91% | $0.15 | 20,104 | |
| none | mini-swe-agent | 0.30% | $0.51 | 159,709 | |
| none | mini-swe-agent | 0% | $3.06 | 39,308 |
The best score, 27%, costs $0.86 a task
Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Terminal-Bench 4.1 leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| MiMo-V2.6 Pro | none | mini-swe-agent | 27% | $0.86 | 245,849 |
| none | mini-swe-agent | 21.8% | $14 | 187,695 | |
| high | mini-SWE-agent | 18.2% | $5.54 | 206,232 | |
| MiMo-V2.6 Flash | none | mini-swe-agent | 16.1% | $0.59 | 200,090 |
| xhigh | mini-swe-agent | 13% | $2.92 | 309,788 | |
| high | mini-SWE-agent | 12.4% | $3.82 | 157,982 | |
| none | mini-swe-agent | 7.58% | $3.48 | 234,421 | |
| Beam 1.0 | none | mini-swe-agent | 6.36% | $0.67 | 64,523 |
| none | mini-swe-agent | 0.91% | $0.15 | 20,104 | |
| none | mini-swe-agent | 0.30% | $0.51 | 159,709 | |
| none | mini-swe-agent | 0% | $3.06 | 39,308 |
Number one over time Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.
Sources: Official Terminal-Bench 4.1 leaderboard and labs’ model cards
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Sources: Official Terminal-Bench 4.1 leaderboard and labs’ model cards
Sources: Official Terminal-Bench 4.1 leaderboard and labs’ model cards
One lab put it in a launch post in the last six months
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| Nov 18, 2026 | 65.2 | Terminal-bench 4.1 (Cheating adjusted accuracy) | |
| Nov 12, 2026 | 57 | Terminal-Bench 4.1 (Cheating-adjusted) |
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Nov 18, 2026 | | 65.2 | Terminal-bench 4.1 (Cheating adjusted accuracy) (results table, p.4) | Capabilities |
One lab claim next to the leaderboard
Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. Every benchmark.
| Model | Lab says | Board says | Gap |
|---|---|---|---|
| | 57% Terminal-Bench 4.1 (Cheating-adjusted) | 56.4% Leaderboard, high, Antigravity CLI | +0.6 pts |
Versions
We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.
Sources: Official leaderboards and boards copied by Epoch AI (CC BY 4.0)
More in Terminal agents
- SkillsBench In use 86-87 containerized expert tasks in 11 domains, run with and without Agent Skills. Verifier tests give pass rate and skill lift. GPT-5.5 66.5%
- Terminal-Bench Hard In use Agents complete 44 hard tasks from Terminal-Bench Core in the Terminus 2 harness. Artificial Analysis scores pass@1 averaged over 3 repeats.
Gemma 4 31B IT 36%
- Harbor-Index In use Agents solve about 80 hard tasks drawn from 29 benchmarks run through Harbor adapters, such as HLE, GSO and SWE-bench Pro. Each task keeps its source verifier. GPT-5.5 28.1%
- Terminal-Bench 2.1 Extended Not in model cards Agents complete long command-line tasks in containers, scored by verification tests. Mercor adds 99 new expert-written tasks to the 89 public ones and reports both scores.
Muse Wonder 54%
- Terminal-Bench Pro Not in model cards Agents solve 400 terminal tasks across eight domains. Harbor distributes the 200 public tasks. Alibaba keeps the other 200 for private evaluation. Claude Opus 4.5 51%
- o11y-bench Not in model cards Agents query metrics, logs and traces in a live Grafana stack, investigate incidents and fix dashboards. Grading checks claims against facts read from the same environment. Claude Opus 5.5 93.8%
Data audit
All 12 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 5 pass
- Pass Independent runs Scores from the official leaderboard (52 entries), launch posts (2 scores) and system cards (1 score).
- Pass Models scored 43 models we track, plus 10 other entries on its leaderboard.
- Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
- Pass Frontier models 10 of the top 10 have a score, led by GPT-6.2 Astra (#1).
- Pass Headroom The best is 71.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on tbench.ai: 65 rows for models we track at the last read.
- Pass Last read November 18th, 2026.
- Pass Read worked The last read worked.
Consistency 2 pass
- Pass Lab claims The one pair agrees within 10 points.
- Pass Known issues No published audit found problems.