Benchmark Almanac

Terminal-Bench 4.0

Claude Opus 5.5 64.8% leads since September 2026 MiMo-V2.6 Pro 27% leads in mini-SWE-agent
Claude Sonnet 5.5 70.6% leads since September 2026
In use
Version 4.0

The official leaderboard hasn’t scored a model released after GPT-6.1 Terra (October 2026). Labs report their own numbers for 9 newer models, such as Claude Fable 5.2 at 67.6%.

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 removed 8 saturated or flawed tasks from Terminal-Bench 3.0 and set a flat 8-hour agent timeout.

Released
August 2026
Kind
Terminal tasks
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Code → Code, Actions, Text
Interaction
Agentic
Environment
Terminal
Items from
Another benchmark
Run by
Anyone
In views
Agents
Strongest model tested
Claude Fable 5.2 #2 on the capability index
Last in a launch post
November 2026
Last model added
November 2026
Scores refreshed
November 20th, 2026
44
Entries on its own leaderboard, 34 models we track
29
Launch posts list it, from 18 labs
15
System cards report it, from 7 labs

Claude Opus 5.5 leads with 64.8% Claude Opus 5.5 leads with 64.8% MiMo-V2.6 Pro leads with 27% in mini-SWE-agent

Best score per entry from the benchmark's own leaderboard, read on November 20th, 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.

Claude Sonnet 5.5 leads with 70.6%, 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.

  1. 1 Claude Opus 5.5 max 64.8%
  2. 2 Claude Sonnet 5.5 max 61.8%
  3. 3 Claude Code + Kimi K3.1 Pro max 60.9%
  4. 4 GPT-6 Astra max 58.2%
  5. 5 GPT-6.1 Sol max 58.2%
  6. 6 Claude Fable 5.1 max 57.9%
  7. 7 Gemini 4 Argon high 54.9%
  8. 8 Claude Opus 5 xhigh 53.9%
  9. 9 GPT-6.1 Terra max 51.8%
  10. 10 GPT-6 Sol max 49.4%
  11. 11 Muse Wonder xhigh 47%
  12. 12 Claude Fable 5 max 44.5%
  13. 13 Claude Code + GLM-5.3 max 41.8%
  14. 14 Claude Haiku 5.5 max 40%
  15. 15 Claude Code + DeepSeek V4.1 Flash 1017 max 40%
  16. 16 Claude Code + Tencent Hy4 max 39.1%
  17. 17 Grok 4.7 xhigh 37.6%
  18. 18 GPT-5.6 Sol max 37.3%
  19. 19 Claude Code + GLM-5.3-Flash max 36.7%
  20. 20 Claude Code + Mistral Large 4 max 27.9%
  21. 21 Claude Code + Qwen3.8 Max (0902) max 27%
  22. 22 MiMo-V2.6 Pro none 27%
  23. 23 Claude Opus 4.8 max 23.6%
  24. 24 Claude Code + Kimi K3 max 22.4%
  25. 25 Ember-1 none 21.8%
  26. 26 GPT-5.6 Terra max 21.5%
  27. 27 Grok 4.6 high 20.3%
  28. 28 Claude Code + Tencent Hy4 preview max 19.4%
  29. 29 Gemini 3.8 Flash high 19.1%
  30. 30 GPT-5.6 Luna max 17.3%
  31. 31 GPT-6 Luna max 16.4%
  32. 32 MiMo-V2.6 Flash none 16.1%
  33. 33 Muse Spark 1.3 xhigh 14.6%
  34. 34 Claude Sonnet 5 max 12.4%
  35. 35 Grok 4.5 high 12.4%
  36. 36 Gemini 3.7 Flash high 11.2%
  37. 37 MiniMax-M3 none 7.27%
  38. 38 Claude Code + DeepSeek V4 Pro 0813 max 6.67%
  39. 39 Beam 1.0 none 6.36%
  40. 40 Gemini 3.1 Pro high 2.73%
  41. 41 GPT-5.4 Mini xhigh 1.82%
  42. 42 Inkling-Small none 0.91%
  43. 43 Nemotron 3.5 Super VL none 0.30%
  44. 44 Inkling none 0%

Source: Official Terminal-Bench 4.0 leaderboard

Source: Official Terminal-Bench 4.0 leaderboard

Sources: Labs’ model cards and official Terminal-Bench 4.0 leaderboard

#EntryScore ± 95% CISettingAgent byReleasedRun onSource
1 Claude Opus 5.5 64.8% ±3.1 max, Claude Code Anthropic Sep 2026 Sep 2026 Official leaderboard
2 Claude Sonnet 5.5 61.8% ±2.9 max, Claude Code Anthropic Sep 2026 Sep 2026 Official leaderboard
3 Claude Code + Kimi K3.1 Pro 60.9% ±4.1 max Anthropic Oct 2026 Oct 2026 Official leaderboard
4 GPT-6 Astra 58.2% ±2.8 max, Codex OpenAI Sep 2026 Sep 2026 Official leaderboard
5 GPT-6.1 Sol 58.2% ±3.1 max, Codex OpenAI Sep 2026 Sep 2026 Official leaderboard
6 Claude Fable 5.1 57.9% ±3.8 max, Claude Code Anthropic Sep 2026 Sep 2026 Official leaderboard
7 Gemini 4 Argon 54.9% ±3 high, Antigravity CLI Google Sep 2026 Sep 2026 Official leaderboard
8 Claude Opus 5 53.9% ±3.2 xhigh, Claude Code Anthropic Jul 2026 Jul 2026 Official leaderboard
9 GPT-6.1 Terra 51.8% ±2.8 max, Codex OpenAI Oct 2026 Oct 2026 Official leaderboard
10 GPT-6 Sol 49.4% ±3.2 max, Codex OpenAI Sep 2026 Sep 2026 Official leaderboard
11 Muse Wonder 47% ±3.8 xhigh, Muse Code Meta Oct 2026 Oct 2026 Official leaderboard
12 Claude Fable 5 44.5% ±3.8 max, Claude Code Anthropic Jun 2026 Jun 2026 Official leaderboard
13 Claude Code + GLM-5.3 41.8% ±3.2 max Anthropic Aug 2026 Aug 2026 Official leaderboard
14 Claude Haiku 5.5 40% ±3.7 max, Claude Code Anthropic Oct 2026 Oct 2026 Official leaderboard
15 Claude Code + DeepSeek V4.1 Flash 1017 40% ±3.7 max Anthropic Oct 2026 Oct 2026 Official leaderboard
16 Claude Code + Tencent Hy4 39.1% ±3.5 max Anthropic Nov 2026 Nov 2026 Official leaderboard
17 Grok 4.7 37.6% ±3.5 xhigh, Grok Build xAI Sep 2026 Sep 2026 Official leaderboard
18 GPT-5.6 Sol 37.3% ±3.8 max, Codex OpenAI Jul 2026 Jun 2026 Official leaderboard
19 Claude Code + GLM-5.3-Flash 36.7% ±3.7 max Anthropic Aug 2026 Aug 2026 Official leaderboard
20 Claude Code + Mistral Large 4 27.9% ±3.9 max Anthropic Oct 2026 Oct 2026 Official leaderboard
21 Claude Code + Qwen3.8 Max (0902) 27% ±3.8 max Anthropic Sep 2026 Sep 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 Claude Code + Kimi K3 22.4% ±3.8 max Anthropic Jul 2026 Jul 2026 Official leaderboard
25 Ember-1 21.8% ±3.4 none, mini-swe-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
26 GPT-5.6 Terra 21.5% ±3.3 max, Codex OpenAI Jul 2026 Jun 2026 Official leaderboard
27 Grok 4.6 20.3% ±3.1 high, Grok Build xAI Aug 2026 Aug 2026 Official leaderboard
28 Claude Code + Tencent Hy4 preview 19.4% ±3.1 max Anthropic Aug 2026 Aug 2026 Official leaderboard
29 Gemini 3.8 Flash 19.1% ±3.4 high, mini-SWE-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
30 GPT-5.6 Luna 17.3% ±2.9 max, Codex OpenAI Jul 2026 Jun 2026 Official leaderboard
31 GPT-6 Luna 16.4% ±2.7 max, Codex OpenAI Sep 2026 Sep 2026 Official leaderboard
32 MiMo-V2.6 Flash 16.1% ±3.4 none, mini-swe-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
33 Muse Spark 1.3 14.6% ±2.9 xhigh, Muse Code Meta Sep 2026 Sep 2026 Official leaderboard
34 Claude Sonnet 5 12.4% ±3.1 max, Claude Code Anthropic Jun 2026 Jun 2026 Official leaderboard
35 Grok 4.5 12.4% ±2.6 high, Grok Build xAI Jul 2026 Jul 2026 Official leaderboard
36 Gemini 3.7 Flash 11.2% ±2.5 high, mini-SWE-agent SWE-agent Aug 2026 Aug 2026 Official leaderboard
37 MiniMax-M3 7.27% ±2.7 none, mini-swe-agent SWE-agent Jun 2026 Jun 2026 Official leaderboard
38 Claude Code + DeepSeek V4 Pro 0813 6.67% ±2.5 max Anthropic Aug 2026 Aug 2026 Official leaderboard
39 Beam 1.0 6.36% ±1.7 none, mini-swe-agent SWE-agent Oct 2026 Oct 2026 Official leaderboard
40 Gemini 3.1 Pro 2.73% ±1.7 high, Antigravity CLI Google Feb 2026 Feb 2026 Official leaderboard
41 GPT-5.4 Mini 1.82% ±1.3 xhigh, Codex OpenAI Mar 2026 Mar 2026 Official leaderboard
42 Inkling-Small 0.91% ±0.9 none, mini-swe-agent SWE-agent Jul 2026 Jul 2026 Official leaderboard
43 Nemotron 3.5 Super VL 0.30% ±0.6 none, mini-swe-agent SWE-agent Oct 2026 Oct 2026 Official leaderboard
44 Inkling 0% none, mini-swe-agent SWE-agent Jul 2026 Jul 2026 Official leaderboard
#ModelScore ± 95% CISettingAgent byReleasedRun onSource
1 Claude Opus 5.5 64.8% ±3.1 max, Claude Code Anthropic Sep 2026 Sep 2026 Official leaderboard
2 Claude Sonnet 5.5 61.8% ±2.9 max, Claude Code Anthropic Sep 2026 Sep 2026 Official leaderboard
3 GPT-6 Astra 58.2% ±2.8 max, Codex OpenAI Sep 2026 Sep 2026 Official leaderboard
4 GPT-6.1 Sol 58.2% ±3.1 max, Codex OpenAI Sep 2026 Sep 2026 Official leaderboard
5 Claude Fable 5.1 57.9% ±3.8 max, Claude Code Anthropic Sep 2026 Sep 2026 Official leaderboard
6 Gemini 4 Argon 54.9% ±3 high, Antigravity CLI Google Sep 2026 Sep 2026 Official leaderboard
7 Claude Opus 5 53.9% ±3.2 xhigh, Claude Code Anthropic Jul 2026 Jul 2026 Official leaderboard
8 GPT-6.1 Terra 51.8% ±2.8 max, Codex OpenAI Oct 2026 Oct 2026 Official leaderboard
9 GPT-6 Sol 49.4% ±3.2 max, Codex OpenAI Sep 2026 Sep 2026 Official leaderboard
10 Muse Wonder 47% ±3.8 xhigh, Muse Code Meta Oct 2026 Oct 2026 Official leaderboard
11 Claude Fable 5 44.5% ±3.8 max, Claude Code Anthropic Jun 2026 Jun 2026 Official leaderboard
12 Claude Haiku 5.5 40% ±3.7 max, Claude Code Anthropic Oct 2026 Oct 2026 Official leaderboard
13 Grok 4.7 37.6% ±3.5 xhigh, Grok Build xAI Sep 2026 Sep 2026 Official leaderboard
14 GPT-5.6 Sol 37.3% ±3.8 max, Codex OpenAI Jul 2026 Jun 2026 Official leaderboard
15 MiMo-V2.6 Pro 27% ±3.6 none, mini-swe-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
16 Claude Opus 4.8 23.6% ±3.6 max, Claude Code Anthropic May 2026 May 2026 Official leaderboard
17 Ember-1 21.8% ±3.4 none, mini-swe-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
18 GPT-5.6 Terra 21.5% ±3.3 max, Codex OpenAI Jul 2026 Jun 2026 Official leaderboard
19 Grok 4.6 20.3% ±3.1 high, Grok Build xAI Aug 2026 Aug 2026 Official leaderboard
20 Gemini 3.8 Flash 19.1% ±3.4 high, mini-SWE-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
21 GPT-5.6 Luna 17.3% ±2.9 max, Codex OpenAI Jul 2026 Jun 2026 Official leaderboard
22 GPT-6 Luna 16.4% ±2.7 max, Codex OpenAI Sep 2026 Sep 2026 Official leaderboard
23 MiMo-V2.6 Flash 16.1% ±3.4 none, mini-swe-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
24 Muse Spark 1.3 14.6% ±2.9 xhigh, Muse Code Meta Sep 2026 Sep 2026 Official leaderboard
25 Claude Sonnet 5 12.4% ±3.1 max, Claude Code Anthropic Jun 2026 Jun 2026 Official leaderboard
26 Grok 4.5 12.4% ±2.6 high, Grok Build xAI Jul 2026 Jul 2026 Official leaderboard
27 Gemini 3.7 Flash 11.2% ±2.5 high, mini-SWE-agent SWE-agent Aug 2026 Aug 2026 Official leaderboard
28 MiniMax-M3 7.27% ±2.7 none, mini-swe-agent SWE-agent Jun 2026 Jun 2026 Official leaderboard
29 Beam 1.0 6.36% ±1.7 none, mini-swe-agent SWE-agent Oct 2026 Oct 2026 Official leaderboard
30 Gemini 3.1 Pro 2.73% ±1.7 high, Antigravity CLI Google Feb 2026 Feb 2026 Official leaderboard
31 GPT-5.4 Mini 1.82% ±1.3 xhigh, Codex OpenAI Mar 2026 Mar 2026 Official leaderboard
32 Inkling-Small 0.91% ±0.9 none, mini-swe-agent SWE-agent Jul 2026 Jul 2026 Official leaderboard
33 Nemotron 3.5 Super VL 0.30% ±0.6 none, mini-swe-agent SWE-agent Oct 2026 Oct 2026 Official leaderboard
34 Inkling 0% none, mini-swe-agent SWE-agent Jul 2026 Jul 2026 Official leaderboard
#ModelScore ± 95% CISettingAgent byReleasedRun onSource
1 MiMo-V2.6 Pro 27% ±3.6 none, mini-swe-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
2 Ember-1 21.8% ±3.4 none, mini-swe-agent SWE-agent Sep 2026 Sep 2026 Official leaderboard
3 Gemini 3.8 Flash 19.1% ±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 Gemini 3.7 Flash 11.2% ±2.5 high, mini-SWE-agent SWE-agent Aug 2026 Aug 2026 Official leaderboard
6 MiniMax-M3 7.27% ±2.7 none, mini-swe-agent SWE-agent Jun 2026 Jun 2026 Official leaderboard
7 Beam 1.0 6.36% ±1.7 none, mini-swe-agent SWE-agent Oct 2026 Oct 2026 Official leaderboard
8 Inkling-Small 0.91% ±0.9 none, mini-swe-agent SWE-agent Jul 2026 Jul 2026 Official leaderboard
9 Nemotron 3.5 Super VL 0.30% ±0.6 none, mini-swe-agent SWE-agent Oct 2026 Oct 2026 Official leaderboard
10 Inkling 0% none, mini-swe-agent SWE-agent Jul 2026 Jul 2026 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Sonnet 5.5 70.6% Sep 2026 Lab model card
2 Claude Fable 5.2 67.6% max effort Nov 2026 Lab model card
3 Claude Opus 5.5 66.4% xhigh effort Sep 2026 Lab model card
4 Kimi K3.1 Pro 65.7% Oct 2026 Lab model card
5 Gemini 4 Carbon 65.2% Oct 2026 Lab model card
6 GPT-6 Astra 58.2% max, Codex Sep 2026 Official leaderboard
7 GPT-6.1 Sol 58.2% max, Codex Sep 2026 Official leaderboard
8 Claude Fable 5.1 57.9% max, Claude Code Sep 2026 Official leaderboard
9 Gemini 4 Argon 57.4% Oct 2026 Lab model card
10 Muse Wonder Full 57.1% Oct 2026 Lab model card
11 Claude Opus 5 53.9% xhigh, Claude Code Jul 2026 Official leaderboard
12 Muse Wonder 52% Oct 2026 Lab model card
13 GPT-6.1 Terra 51.8% max, Codex Oct 2026 Official leaderboard
14 GPT-6 Sol 49.4% max, Codex Sep 2026 Official leaderboard
15 Qwen4-Max 48.5% xhigh effort Nov 2026 Lab model card
16 Ling-3.1-flash 46% Oct 2026 Lab model card
17 Claude Fable 5 44.5% max, Claude Code Jun 2026 Official leaderboard
18 Tencent Hy4 44.4% Nov 2026 Lab model card
19 DeepSeek V4.1 Flash 1017 42.3% Oct 2026 Lab model card
20 Mistral Large 4 (Le Chonk, open weights) 41% Oct 2026 Lab model card
21 Muse Ember 40.9% max effort Nov 2026 Lab model card
22 Claude Haiku 5.5 40% max, Claude Code Oct 2026 Official leaderboard
23 LongCat-2.5 38.5% Nov 2026 Lab model card
24 MiniMax-M3.1-Flash 38.1% Oct 2026 Lab model card
25 Grok 4.7 38% Sep 2026 Lab model card
26 GPT-5.6 Sol 37.3% max, Codex Jul 2026 Official leaderboard
27 MiMo-V2.6 Pro 34.9% Sep 2026 Lab model card
28 Step 5 Preview 33.3% Sep 2026 Lab model card
29 DeepSeek V4.1 Flash 31.2% Sep 2026 Lab model card
30 MiMo-V2.6 Flash 28.8% Sep 2026 Lab model card
31 Mistral Large 4 28.3% Oct 2026 Lab model card
32 SWE-2 27.3% Sep 2026 Lab model card
33 Claude Opus 4.8 23.6% max, Claude Code May 2026 Official leaderboard
34 Ember-1 21.8% none, mini-swe-agent Sep 2026 Official leaderboard
35 GPT-5.6 Terra 21.5% max, Codex Jul 2026 Official leaderboard
36 Grok 4.6 20.3% high, Grok Build Aug 2026 Official leaderboard
37 SWE-2 Flash 20% Nov 2026 Lab model card
38 Gemini 3.8 Flash 19.1% Sep 2026 Lab model card
39 GPT-5.6 Luna 17.3% max, Codex Jul 2026 Official leaderboard
40 GPT-6 Luna 16.4% max, Codex Sep 2026 Official leaderboard
41 Muse Spark 1.3 14.6% xhigh, Muse Code Sep 2026 Official leaderboard
42 Claude Sonnet 5 12.4% max, Claude Code Jun 2026 Official leaderboard
43 Grok 4.5 12.4% high, Grok Build Jul 2026 Official leaderboard
44 Gemini 3.7 Flash 11.2% high, mini-SWE-agent Aug 2026 Official leaderboard
45 MiniMax-M3 7.27% none, mini-swe-agent Jun 2026 Official leaderboard
46 Beam 1.0 6.36% none, mini-swe-agent Oct 2026 Official leaderboard
47 Inkling 2 Mini 5.60% Nov 2026 Lab model card
48 Gemini 3.1 Pro 2.73% high, Antigravity CLI Feb 2026 Official leaderboard
49 GPT-5.4 Mini 1.82% xhigh, Codex Mar 2026 Official leaderboard
50 Nemotron 3.5 Super VL 1.00% Oct 2026 Lab model card
51 Inkling-Small 0.91% none, mini-swe-agent Jul 2026 Official leaderboard
52 Inkling 0% none, mini-swe-agent Jul 2026 Official leaderboard

Labs report scores for 31 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.

0% 20% 40% 60% 80% 1001,00010K Mean time per trial (s), log scale, lower is better Accuracy Claude Opus 5.5 (max)Claude Code + Kimi K3.1 Pro (max)GPT-6 Astra (max)GPT-6 Astra (high)GPT-6 Astra (medium)GPT-6 Astra (low)Beam 1.0 (none)Inkling-Small (none) 0% 20% 40% 60% 80% 1001,00010K Mean time per trial (s), log scale, lower is better Accuracy Claude Opus 5.5 (max)Claude Code + Kimi K3.1 Pro (max)GPT-6 Astra (max)GPT-6 Astra (high)GPT-6 Astra (medium)GPT-6 Astra (low)Beam 1.0 (none)Inkling-Small (none)
ModelSettingAccuracyMean time per trial
Claude Opus 5.5 max, Claude Code 64.9% 4,790 s
Claude Sonnet 5.5 max, Claude Code 61.8% 7,432 s
Claude Code + Kimi K3.1 Pro max, Claude Code 60.9% 4,047 s
GPT-6 Astra max, Codex 58.2% 2,796 s
GPT-6.1 Sol max, Codex 58.2% 3,177 s
GPT-6 Astra high, Codex 57.9% 2,114 s
GPT-6 Astra xhigh, Codex 57.9% 2,209 s
Claude Fable 5.1 xhigh, Claude Code 57.9% 3,202 s
Claude Fable 5.1 max, Claude Code 57.9% 3,894 s
Gemini 4 Argon high, Antigravity CLI 54.9% 3,895 s
Claude Fable 5.1 high, Claude Code 54.6% 3,018 s
GPT-6 Astra medium, Codex 54.2% 1,886 s
Claude Fable 5.1 medium, Claude Code 53.9% 2,528 s
Claude Opus 5 xhigh, Claude Code 53.9% 4,515 s
GPT-6.1 Terra max, Codex 51.8% 2,291 s
Claude Opus 5 max, Claude Code 51.8% 4,793 s
GPT-6 Astra low, Codex 50.6% 1,684 s
Claude Opus 5 high, Claude Code 50.3% 3,822 s
GPT-6 Sol max, Codex 49.4% 2,662 s
Muse Wonder xhigh, Muse Code 47% 2,828 s
Claude Opus 5 medium, Claude Code 44.9% 3,293 s
Claude Fable 5 max, Claude Code 44.6% 4,203 s
Claude Fable 5.1 low, Claude Code 43.3% 2,291 s
Claude Code + GLM-5.3 max, Claude Code 41.8% 5,831 s
Claude Haiku 5.5 max, Claude Code 40% 4,059 s
Claude Code + DeepSeek V4.1 Flash 1017 max, Claude Code 40% 5,030 s
Claude Code + Tencent Hy4 max, Claude Code 39.1% 4,882 s
Grok 4.7 xhigh, Grok Build 37.6% 5,701 s
GPT-5.6 Sol max, Codex 37.3% 2,388 s
Claude Code + GLM-5.3-Flash max, Claude Code 36.7% 6,558 s
Claude Code + GLM-5.3-Flash none, Claude Code 35.8% 6,210 s
Claude Opus 5 low, Claude Code 34.9% 2,947 s
Claude Code + Mistral Large 4 max, Claude Code 27.9% 5,229 s
MiMo-V2.6 Pro none, mini-swe-agent 27% 5,265 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 22.4% 3,351 s
Ember-1 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 20.3% 2,352 s
Claude Code + Tencent Hy4 preview max, Claude Code 19.4% 6,967 s
Gemini 3.8 Flash high, mini-SWE-agent 19.1% 1,995 s
GPT-5.6 Luna max, Codex 17.3% 4,088 s
GPT-6 Luna max, Codex 16.4% 2,284 s
MiMo-V2.6 Flash none, mini-swe-agent 16.1% 5,472 s
Muse Spark 1.3 xhigh, Muse Code 14.6% 4,342 s
Grok 4.5 high, Grok Build 12.4% 3,162 s
Claude Sonnet 5 max, Claude Code 12.4% 6,510 s
Gemini 3.7 Flash high, mini-SWE-agent 11.2% 1,689 s
MiniMax-M3 none, mini-swe-agent 7.3% 3,507 s
Claude Code + DeepSeek V4 Pro 0813 max, Claude Code 6.7% 3,140 s
Beam 1.0 none, mini-swe-agent 6.4% 1,440 s
Gemini 3.1 Pro high, Antigravity CLI 2.7% 4,583 s
GPT-5.4 Mini xhigh, Codex 1.8% 1,849 s
Inkling-Small none, mini-swe-agent 0.9% 475.1 s
Nemotron 3.5 Super VL none, mini-swe-agent 0.3% 1,781 s
Inkling none, mini-swe-agent 0% 1,403 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.

0% 20% 40% 60% 80% 1001,00010K Mean time per trial (s), log scale, lower is better Accuracy Claude Opus 5.5 (max)GPT-6 Astra (max)GPT-6 Astra (high)GPT-6 Astra (medium)GPT-6 Astra (low)Beam 1.0 (none)Inkling-Small (none) 0% 20% 40% 60% 80% 1001,00010K Mean time per trial (s), log scale, lower is better Accuracy Claude Opus 5.5 (max)GPT-6 Astra (max)GPT-6 Astra (high)GPT-6 Astra (medium)GPT-6 Astra (low)Beam 1.0 (none)Inkling-Small (none)
ModelSettingAccuracyMean time per trial
Claude Opus 5.5 max, Claude Code 64.9% 4,790 s
Claude Sonnet 5.5 max, Claude Code 61.8% 7,432 s
GPT-6 Astra max, Codex 58.2% 2,796 s
GPT-6.1 Sol max, Codex 58.2% 3,177 s
GPT-6 Astra high, Codex 57.9% 2,114 s
GPT-6 Astra xhigh, Codex 57.9% 2,209 s
Claude Fable 5.1 xhigh, Claude Code 57.9% 3,202 s
Claude Fable 5.1 max, Claude Code 57.9% 3,894 s
Gemini 4 Argon high, Antigravity CLI 54.9% 3,895 s
Claude Fable 5.1 high, Claude Code 54.6% 3,018 s
GPT-6 Astra medium, Codex 54.2% 1,886 s
Claude Fable 5.1 medium, Claude Code 53.9% 2,528 s
Claude Opus 5 xhigh, Claude Code 53.9% 4,515 s
GPT-6.1 Terra max, Codex 51.8% 2,291 s
Claude Opus 5 max, Claude Code 51.8% 4,793 s
GPT-6 Astra low, Codex 50.6% 1,684 s
Claude Opus 5 high, Claude Code 50.3% 3,822 s
GPT-6 Sol max, Codex 49.4% 2,662 s
Muse Wonder xhigh, Muse Code 47% 2,828 s
Claude Opus 5 medium, Claude Code 44.9% 3,293 s
Claude Fable 5 max, Claude Code 44.6% 4,203 s
Claude Fable 5.1 low, Claude Code 43.3% 2,291 s
Claude Haiku 5.5 max, Claude Code 40% 4,059 s
Grok 4.7 xhigh, Grok Build 37.6% 5,701 s
GPT-5.6 Sol max, Codex 37.3% 2,388 s
Claude Opus 5 low, Claude Code 34.9% 2,947 s
MiMo-V2.6 Pro none, mini-swe-agent 27% 5,265 s
Claude Opus 4.8 max, Claude Code 23.6% 5,163 s
Ember-1 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 20.3% 2,352 s
Gemini 3.8 Flash high, mini-SWE-agent 19.1% 1,995 s
GPT-5.6 Luna max, Codex 17.3% 4,088 s
GPT-6 Luna max, Codex 16.4% 2,284 s
MiMo-V2.6 Flash none, mini-swe-agent 16.1% 5,472 s
Muse Spark 1.3 xhigh, Muse Code 14.6% 4,342 s
Grok 4.5 high, Grok Build 12.4% 3,162 s
Claude Sonnet 5 max, Claude Code 12.4% 6,510 s
Gemini 3.7 Flash high, mini-SWE-agent 11.2% 1,689 s
MiniMax-M3 none, mini-swe-agent 7.3% 3,507 s
Beam 1.0 none, mini-swe-agent 6.4% 1,440 s
Gemini 3.1 Pro high, Antigravity CLI 2.7% 4,583 s
GPT-5.4 Mini xhigh, Codex 1.8% 1,849 s
Inkling-Small none, mini-swe-agent 0.9% 475.1 s
Nemotron 3.5 Super VL none, mini-swe-agent 0.3% 1,781 s
Inkling none, mini-swe-agent 0% 1,403 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.

0% 10% 20% 30% 1001,00010K Mean time per trial (s), log scale, lower is better Accuracy MiMo-V2.6 Pro (none)Ember-1 (none)Gemini 3.8 Flash (high)Gemini 3.7 Flash (high)Beam 1.0 (none)Inkling-Small (none) 0% 10% 20% 30% 1001,00010K Mean time per trial (s), log scale, lower is better Accuracy MiMo-V2.6 Pro (none)Ember-1 (none)Gemini 3.8 Flash (high)Gemini 3.7 Flash (high)Beam 1.0 (none)Inkling-Small (none)
ModelSettingAccuracyMean time per trial
MiMo-V2.6 Pro none, mini-swe-agent 27% 5,265 s
Ember-1 none, mini-swe-agent 21.8% 4,960 s
Gemini 3.8 Flash high, mini-SWE-agent 19.1% 1,995 s
MiMo-V2.6 Flash none, mini-swe-agent 16.1% 5,472 s
Gemini 3.7 Flash high, mini-SWE-agent 11.2% 1,689 s
MiniMax-M3 none, mini-swe-agent 7.3% 3,507 s
Beam 1.0 none, mini-swe-agent 6.4% 1,440 s
Inkling-Small none, mini-swe-agent 0.9% 475.1 s
Nemotron 3.5 Super VL none, mini-swe-agent 0.3% 1,781 s
Inkling none, mini-swe-agent 0% 1,403 s

The best score, 64.8%, costs $14 a task

Runs on the cost frontier score higher than any run that's cheaper. Claude Code + Kimi K3.1 Pro at max comes within 3.9 points of the best for $3.29 a task, 4 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

Thinking Machines Anthropic OpenAI Agents other companies built, in a pale shade of the model’s lab Cost frontier One model at each effort level
0% 20% 40% 60% 80% $0.10 $1.00 $10 $100 Cost per task, log scale Claude Opus 5.5, max Claude Code + Kimi K3.1 Pro, max Claude Code + DeepSeek V4.1 Flash 1017, max Claude Code + GLM-5.3-Flash, none GPT-6 Luna, max Inkling-Small, none 0% 20% 40% 60% 80% $0.10 $1.00 $10 $100 Cost per task, log scale Claude Opus 5.5, max GPT-6.1 Sol, max

Source: Official Terminal-Bench 4.0 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 max Claude Code 64.8% $14 269,135
Claude Sonnet 5.5 max Claude Code 61.8% $22 565,388
Claude Code + Kimi K3.1 Pro max Claude Code 60.9% $3.29 80,956
GPT-6 Astra max Codex 58.2% $9.90 72,694
GPT-6.1 Sol max Codex 58.2% $1.92 94,892
Claude Fable 5.1 max Claude Code 57.9% $19 191,277
Claude Fable 5.1 xhigh Claude Code 57.9% $15 152,809
GPT-6 Astra xhigh Codex 57.9% $7.12 45,304
GPT-6 Astra high Codex 57.9% $6.88 41,527
Gemini 4 Argon high Antigravity CLI 54.9% $3.36 162,988
Claude Fable 5.1 high Claude Code 54.5% $12 113,421
GPT-6 Astra medium Codex 54.2% $5.80 32,638
Claude Opus 5 xhigh Claude Code 53.9% $18 179,281
Claude Fable 5.1 medium Claude Code 53.9% $8.59 78,763
Claude Opus 5 max Claude Code 51.8% $18 200,037
GPT-6.1 Terra max Codex 51.8% $2.65 100,138
GPT-6 Astra low Codex 50.6% $4.72 24,207
Claude Opus 5 high Claude Code 50.3% $14 144,242
GPT-6 Sol max Codex 49.4% $4.40 92,642
Muse Wonder xhigh Muse Code 47% $7.43 129,324
Claude Opus 5 medium Claude Code 44.9% $9.67 101,450
Claude Fable 5 max Claude Code 44.5% $22 177,636
Claude Fable 5.1 low Claude Code 43.3% $7.15 60,409
Claude Code + GLM-5.3 max Claude Code 41.8% $8.27 208,061
Claude Haiku 5.5 max Claude Code 40% $2.89 386,483
Claude Code + DeepSeek V4.1 Flash 1017 max Claude Code 40% $0.56 497,265
Claude Code + Tencent Hy4 max Claude Code 39.1% $1.05 368,811
Grok 4.7 xhigh Grok Build 37.6% $11 69,627
GPT-5.6 Sol max Codex 37.3% $7.70 71,065
Claude Code + GLM-5.3-Flash max Claude Code 36.7% $1.14 459,702
Claude Code + GLM-5.3-Flash none Claude Code 35.8% $0.36 135,014
Claude Opus 5 low Claude Code 34.9% $7.25 72,229
Claude Code + Mistral Large 4 max Claude Code 27.9% $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 22.4% $5.82 195,427
Ember-1 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 20.3% $11 98,843
Claude Code + Tencent Hy4 preview max Claude Code 19.4% $1.99 309,296
Gemini 3.8 Flash high mini-SWE-agent 19.1% $5.54 206,232
GPT-5.6 Luna max Codex 17.3% $1.05 185,705
GPT-6 Luna max Codex 16.4% $0.22 137,851
MiMo-V2.6 Flash none mini-swe-agent 16.1% $0.59 200,090
Muse Spark 1.3 xhigh Muse Code 14.6% 265,112
Claude Sonnet 5 max Claude Code 12.4% $29 353,330
Grok 4.5 high Grok Build 12.4% $6.35 98,149
Gemini 3.7 Flash high mini-SWE-agent 11.2% $3.82 157,982
MiniMax-M3 none mini-swe-agent 7.27% $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% $0.71 64,523
Gemini 3.1 Pro high Antigravity CLI 2.73% $5.97 79,607
GPT-5.4 Mini xhigh Codex 1.82% $2.04 169,597
Inkling-Small none mini-swe-agent 0.91% $0.15 20,104
Nemotron 3.5 Super VL none mini-swe-agent 0.30% $0.51 159,709
Inkling none mini-swe-agent 0% $3.06 39,308

The best score, 64.8%, costs $14 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.

Thinking Machines Xiaomi Corp Anthropic OpenAI Cost frontier One model at each effort level
0% 20% 40% 60% 80% $0.10 $1.00 $10 $100 Cost per task, log scale Claude Opus 5.5, max GPT-6.1 Sol, max MiMo-V2.6 Pro, none GPT-6 Luna, max Inkling-Small, none 0% 20% 40% 60% 80% $0.10 $1.00 $10 $100 Cost per task, log scale Claude Opus 5.5, max GPT-6.1 Sol, max

Source: Official Terminal-Bench 4.0 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 max Claude Code 64.8% $14 269,135
Claude Sonnet 5.5 max Claude Code 61.8% $22 565,388
GPT-6 Astra max Codex 58.2% $9.90 72,694
GPT-6.1 Sol max Codex 58.2% $1.92 94,892
Claude Fable 5.1 max Claude Code 57.9% $19 191,277
Claude Fable 5.1 xhigh Claude Code 57.9% $15 152,809
GPT-6 Astra xhigh Codex 57.9% $7.12 45,304
GPT-6 Astra high Codex 57.9% $6.88 41,527
Gemini 4 Argon high Antigravity CLI 54.9% $3.36 162,988
Claude Fable 5.1 high Claude Code 54.5% $12 113,421
GPT-6 Astra medium Codex 54.2% $5.80 32,638
Claude Opus 5 xhigh Claude Code 53.9% $18 179,281
Claude Fable 5.1 medium Claude Code 53.9% $8.59 78,763
Claude Opus 5 max Claude Code 51.8% $18 200,037
GPT-6.1 Terra max Codex 51.8% $2.65 100,138
GPT-6 Astra low Codex 50.6% $4.72 24,207
Claude Opus 5 high Claude Code 50.3% $14 144,242
GPT-6 Sol max Codex 49.4% $4.40 92,642
Muse Wonder xhigh Muse Code 47% $7.43 129,324
Claude Opus 5 medium Claude Code 44.9% $9.67 101,450
Claude Fable 5 max Claude Code 44.5% $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 37.6% $11 69,627
GPT-5.6 Sol max Codex 37.3% $7.70 71,065
Claude Opus 5 low Claude Code 34.9% $7.25 72,229
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
Ember-1 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 20.3% $11 98,843
Gemini 3.8 Flash high mini-SWE-agent 19.1% $5.54 206,232
GPT-5.6 Luna max Codex 17.3% $1.05 185,705
GPT-6 Luna max Codex 16.4% $0.22 137,851
MiMo-V2.6 Flash none mini-swe-agent 16.1% $0.59 200,090
Muse Spark 1.3 xhigh Muse Code 14.6% 265,112
Claude Sonnet 5 max Claude Code 12.4% $29 353,330
Grok 4.5 high Grok Build 12.4% $6.35 98,149
Gemini 3.7 Flash high mini-SWE-agent 11.2% $3.82 157,982
MiniMax-M3 none mini-swe-agent 7.27% $3.48 234,421
Beam 1.0 none mini-swe-agent 6.36% $0.71 64,523
Gemini 3.1 Pro high Antigravity CLI 2.73% $5.97 79,607
GPT-5.4 Mini xhigh Codex 1.82% $2.04 169,597
Inkling-Small none mini-swe-agent 0.91% $0.15 20,104
Nemotron 3.5 Super VL none mini-swe-agent 0.30% $0.51 159,709
Inkling 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.

Thinking Machines Xiaomi Corp Cost frontier One model at each effort level
0% 10% 20% 30% $0.10 $1.00 $10 $100 Cost per task, log scale MiMo-V2.6 Pro, none MiMo-V2.6 Flash, none Inkling-Small, none 0% 10% 20% 30% $0.10 $1.00 $10 $100 Cost per task, log scale MiMo-V2.6 Pro, none MiMo-V2.6 Flash, none

Source: Official Terminal-Bench 4.0 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
MiMo-V2.6 Pro none mini-swe-agent 27% $0.86 245,849
Ember-1 none mini-swe-agent 21.8% $14 187,695
Gemini 3.8 Flash high mini-SWE-agent 19.1% $5.54 206,232
MiMo-V2.6 Flash none mini-swe-agent 16.1% $0.59 200,090
Gemini 3.7 Flash high mini-SWE-agent 11.2% $3.82 157,982
MiniMax-M3 none mini-swe-agent 7.27% $3.48 234,421
Beam 1.0 none mini-swe-agent 6.36% $0.71 64,523
Inkling-Small none mini-swe-agent 0.91% $0.15 20,104
Nemotron 3.5 Super VL none mini-swe-agent 0.30% $0.51 159,709
Inkling none mini-swe-agent 0% $3.06 39,308

4 models have held number one on the official leaderboard since its release in August 2026 5 models have held number one with claims included since its release in August 2026

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.

Anthropic OpenAI
Own leaderboard 4 leaders
Lab claims 4 leaders

Sources: Official Terminal-Bench 4.0 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.

Plot with other benchmarks
On the benchmark’s own leaderboard Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Sonnet 5.5 Claude Opus 5.5 GPT-6 Astra Claude Opus 5 Claude Fable 5 Claude Opus 4.8 Gemini 3.1 Pro 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Sonnet 5.5 GPT-6 Astra

Sources: Official Terminal-Bench 4.0 leaderboard and labs’ model cards

On the benchmark’s own leaderboard Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Sonnet 5.5 Claude Opus 5.5 GPT-6 Astra Gemini 3.7 Flash MiniMax-M3 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Sonnet 5.5 GPT-6 Astra

Sources: Official Terminal-Bench 4.0 leaderboard and labs’ model cards

18 labs 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.

Google DeepMind
OpenAI
DeepSeek
Cognition
StepFun
xAI
Xiaomi Corp
Anthropic
Mistral AI
Meta
Ant Group
NVIDIA
MiniMax
Moonshot AI
Meituan
Tencent
Qwen
Thinking Machines
DateLaunch postScoreAs the post writes it
Nov 20, 2026 Inkling 2 Mini 5.6 Terminal Bench v4.0 (mini-swe-agent v2.4.6)
Nov 15, 2026 Qwen4-Max 48.5 Agentic terminal coding, Terminal-Bench 4.0 (Xhigh effort)
Nov 9, 2026 Muse Ember 40.9 Terminal-Bench 4.0, best published harness, Muse Ember max reasoning
Nov 4, 2026 Tencent Hy4 44.4 Terminal Bench 4.0
Nov 4, 2026 SWE-2 Flash 20 Terminal-Bench 4.0
Nov 2, 2026 LongCat-2.5 38.5 Terminal Bench v4.0
Oct 29, 2026 Mistral Large 4 (Le Chonk, open weights) 41 Terminal-Bench 4.0 (Artificial Analysis Coding Agent Index run)
Oct 28, 2026 Gemini 4 Carbon 65.2 Terminal-Bench 4.0 (AA)
Oct 26, 2026 GPT-6.1 Terra 50.3 Terminal-Bench 4.0, maximum reasoning effort
Oct 24, 2026 Kimi K3.1 Pro 65.7 Terminal-Bench 4.0 (Claude Code)
Oct 19, 2026 MiniMax-M3.1-Flash 38.1 Terminal Bench 4.0
Oct 17, 2026 DeepSeek V4.1 Flash 1017 42.3 Terminal-Bench 4.0
Oct 15, 2026 Nemotron 3.5 Super VL 1 Terminal Bench 4 [offline]
Oct 13, 2026 Ling-3.1-flash 46 Terminal-Bench-4.0 (LingCode)
Oct 13, 2026 Gemini 4 Argon 57.4 Terminal-bench 4.0
Oct 12, 2026 Muse Wonder 52 Terminal-Bench 4.0-AA
Oct 12, 2026 Muse Wonder Full 57.1 Terminal-Bench 4.0-AA
Oct 7, 2026 Claude Haiku 5.5 39.2 Agentic coding Terminal-Bench 4.0
Oct 6, 2026 Mistral Large 4 28.3 Coding - Terminal Bench 4.0: AA - Terminal-Bench 4 (chart 28; text 28.3%)
Sep 28, 2026 Claude Sonnet 5.5 70.6 Agentic coding Terminal-Bench 4.0
Sep 22, 2026 MiMo-V2.6 Pro 34.9 Terminal Bench 4.0
Sep 22, 2026 Claude Opus 5.5 66.4 Agentic coding Terminal-Bench 4.0 (xhigh effort)
Sep 21, 2026 MiMo-V2.6 Flash 28.8 Terminal Bench 4.0
Sep 21, 2026 Grok 4.7 37.6 Terminal-Bench 4.0 (Multi-hour terminal work)
Sep 20, 2026 Step 5 Preview 33.3 Terminal-Bench v4
Sep 10, 2026 SWE-2 27.3 Terminal-Bench 4.0
Sep 10, 2026 DeepSeek V4.1 Flash 31.2 Terminal-Bench 4.0 (Pass@1)
Sep 3, 2026 GPT-6 Astra 57.9 Terminal-Bench 4.0
Sep 2, 2026 Gemini 3.8 Flash 19.1 Terminal-bench 4.0

15 system cards report it, from 7 labs

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.

DateDocumentScoreAs the document writes itSection
Nov 20, 2026 Inkling 2 Mini model card 5.6 Terminal Bench v4.0 (mini-swe-agent v2.4.6) Capabilities
Nov 10, 2026 Gemini 4 Carbon model card 65.2 Terminal-Bench 4.0 (AA) (results table) Capabilities
Nov 9, 2026 Muse Ember technical report 40.9 Terminal-Bench 4.0, best published harness, Muse Ember max reasoning Capabilities
Nov 9, 2026 Claude Fable 5.2 system card 67.6 Terminal-Bench 4.0, safeguards enabled, Claude Code --bare, max effort, no internet access, 10 trials per task. Mythos-only score 70.5% excluded (Section 8.5) Capabilities
Oct 13, 2026 Gemini 4 Argon model card 57.4 Terminal-bench 4.0 (results table, p.5) Capabilities
Oct 12, 2026 Muse Wonder model card 52 Terminal-Bench 4.0-AA accuracy from Artificial Analysis, Muse Wonder (max) (capabilities table image, page 3) Capabilities
Oct 12, 2026 Muse Wonder Full model card 57.1 Terminal-Bench 4.0-AA accuracy from Artificial Analysis, Muse Wonder Full (max) (capabilities table image, page 3) Capabilities
Oct 7, 2026 Claude Haiku 5.5 system card 39.2 Terminal-Bench 4.0, max effort, safeguards on with no fallback, no internet egress (8.4) Capabilities
Sep 28, 2026 Claude Sonnet 5.5 system card 70.6 Terminal-Bench 4.0, max effort with safeguards and fallbacks, five trials per task (8.5) Capabilities
Sep 22, 2026 MiMo-V2.6 Pro technical report 34.9 Terminal Bench 4.0 (Table 3) Capabilities
Sep 22, 2026 Claude Opus 5.5 system card 66.4 Terminal-Bench 4.0 at xhigh effort, the setting the summary table marks (64.8% at max), Claude Code --bare mode, safeguards with fallback, five trials per task (Table 8.1.A, 8.5) Capabilities
Sep 21, 2026 Grok 4.7 model card 38 Terminal-Bench 4.0, task success rate, xhigh, Grok Build harness, run by Harbor (2.3) Capabilities
Sep 10, 2026 DeepSeek V4.1 Flash technical report 31.2 Terminal-Bench 4.0 (Pass@1) (Table 3) Capabilities
Sep 2, 2026 Gemini 3.8 Flash model card 19.1 Terminal-bench 4.0 (results table, p.5) Capabilities
Sep 1, 2026 Claude Fable 5.1 system card 55.8 Terminal-Bench 4.0, Claude Code --bare mode, max effort, 15 trials per task; Table 8.1.A rounds to 56% (Mythos 5.1 61%) (8.6) Capabilities

The lab's number comes out ahead in 8 of 12 pairs that differ

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. 1 of them match, which usually means one side copied the other. Every benchmark.

Model Lab says Board says Gap
MiMo-V2.6 Flash 28.8% Terminal Bench 4.0 16.1% Leaderboard, none, mini-swe-agent +13 pts
Claude Sonnet 5.5 70.6% Agentic coding Terminal-Bench 4.0 61.8% Leaderboard, max, Claude Code +8.8 pts
MiMo-V2.6 Pro 34.9% Terminal Bench 4.0 27% Leaderboard, none, mini-swe-agent +7.9 pts
Muse Wonder 52% Terminal-Bench 4.0-AA 47% Leaderboard, xhigh, Muse Code +5.0 pts
Gemini 4 Argon 57.4% Terminal-bench 4.0 54.9% Leaderboard, high, Antigravity CLI +2.5 pts
Claude Fable 5.1 55.8% Terminal-Bench 4.0, Claude Code --bare mode, max effort, 15 trials per task; Table 8.1.A rounds to 56% (Mythos 5.1 61%) (8.6) 57.9% Leaderboard, max, Claude Code −2.1 pts
Claude Opus 5.5 66.4% Agentic coding Terminal-Bench 4.0 (xhigh effort) 64.8% Leaderboard, max, Claude Code +1.6 pts
GPT-6.1 Terra 50.3% Terminal-Bench 4.0, maximum reasoning effort 51.8% Leaderboard, max, Codex −1.5 pts
Claude Haiku 5.5 39.2% Agentic coding Terminal-Bench 4.0 40% Leaderboard, max, Claude Code −0.8 pts
Nemotron 3.5 Super VL 1.00% Terminal Bench 4 [offline] 0.30% Leaderboard, none, mini-swe-agent +0.7 pts
Grok 4.7 38% Terminal-Bench 4.0, task success rate, xhigh, Grok Build harness, run by Harbor (2.3) 37.6% Leaderboard, xhigh, Grok Build +0.4 pts
GPT-6 Astra 57.9% Terminal-Bench 4.0 58.2% Leaderboard, max, Codex −0.3 pts
Gemini 3.8 Flash 19.1% Terminal-bench 4.0 19.1% Leaderboard, high, mini-SWE-agent Same

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.

Google DeepMind Anthropic OpenAI GitHub
Terminal-Bench 4.1 20 leaders
Copilot Fusion 71.2%
Terminal-Bench Replaced by 2.0, Nov 2025
GPT-5 49.6%
Terminal-Bench 2.0 Replaced by 2.1, May 2026
GPT-5.5 82.3%
Terminal-Bench 2.1 Passed 85% in Sep 2026, by GPT-6 Astra
GPT-6 Astra 87.4%
Terminal-Bench 3.0 Replaced by 4.0, Aug 2026
Copilot Fusion 71.2%

Sources: Official leaderboards and boards copied by Epoch AI (CC BY 4.0)

More in Terminal agents

All 35 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

11 of 12 checks pass, 1 warning. Look first at the Lab claims check. 1 of 13 pairs differs by more than 10 points, the largest for MiMo-V2.6 Flash (+13 pts). 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 (44 entries), launch posts (29 scores) and system cards (15 scores).
  • Pass Models scored 52 models we track, plus 10 other entries on its leaderboard.
  • Pass Newest score The newest is from November 2026, for Inkling 2 Mini.
  • Pass Frontier models 9 of the top 10 have a score, led by Claude Fable 5.2 (#2).
  • Pass Headroom The best is 70.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on tbench.ai: 57 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 warning 1 pass
  • Warning Lab claims 1 of 13 pairs differs by more than 10 points, the largest for MiMo-V2.6 Flash (+13 pts).
  • Pass Known issues No published audit found problems.

How we check