Terminal-Bench 2.1
Agents · 2026-05-06
Terminal-Bench 2.1 is a revision of Terminal-Bench 2.0 that fixes issues in 28 of its 89 command-line tasks and introduces continuous validation. Scores are reported as average task accuracy.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Mythos Preview | 92.1 |
| Claude Opus 5 | 89.1 |
| GPT-5.6 Sol | 88.8 |
| Kimi K3 | 88.3 |
| Claude Mythos 5 | 88.0 |
| Macaron-V1-Venti | 87.6 |
| GPT-5.6 Terra | 87.4 |
| GPT-5.6 Luna | 84.7 |
| Claude Fable 5 | 84.3 |
| GPT-5.5 | 83.4 |
| Grok 4.5 | 83.3 |
| Opus 4.7 | 83.1 |
| Opus 4.8 | 82.7 |
| GLM-5.2 | 82.7 |
| Sonnet 5 | 80.5 |
| Muse Spark 1.1 | 80.0 |
| DeepSeek-V4-Flash-0731 | 78.7 |
| Gemini 3.5 Flash | 78.7 |
| GPT-5.4 | 78.3 |
| Gemini 3.6 Flash | 78.0 |
| Gemini 3 Pro Preview | 73.9 |
| Sonnet 4.6 | 71.2 |
| Gemini 3.1 Pro Preview | 70.8 |
| Motif-3-Beta | 70.8 |
| Laguna S 2.1 | 70.2 |
| KAT-Coder-Pro V2 | 70.0 |
| Nex-N2-Pro | 67.8 |
| Kimi K2.7 Code | 67.4 |
| Agnes 2.5 Pro Alpha | 67.0 |
| MiniMax M3 | 66.0 |
| Kimi K2.6 | 65.9 |
| MiMo-V2.5-Pro | 65.2 |
| DeepSeek-V4-Pro | 64.8 |
| Inkling-Small | 64.7 |
| Inkling | 63.8 |
| MiMo-V2.5 | 63.7 |
| Muse Spark | 62.2 |
| GLM-5.1 | 61.8 |
| DeepSeek-V4-Flash | 61.8 |
| MiMo-V2-Flash | 61.8 |