Terminal-Bench 2.1 - Easy
Agents · 2026-05-06
The Easy split of Terminal-Bench 2.1 reports the percentage of easy terminal-use tasks completed successfully; higher scores are better.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 100.0 |
| Opus 4.8 | 100.0 |
| Claude Opus 5 | 100.0 |
| Sonnet 4.6 | 100.0 |
| Sonnet 5 | 100.0 |
| Composer 2.5 | 100.0 |
| Gemini 3.5 Flash | 100.0 |
| Gemini 3.6 Flash | 100.0 |
| GLM-5.2 | 100.0 |
| MiMo-V2.5 | 100.0 |
| Opus 4.7 | 91.7 |
| Gemini 3.1 Pro Preview | 91.7 |
| GPT-5.6 Sol | 91.7 |
| Grok 4.5 | 91.7 |
| Inkling | 91.7 |
| Kimi K2.7 Code | 91.7 |
| Kimi K3 | 91.7 |
| Nemotron 3 Ultra 550B A55B | 91.7 |
| Haiku 4.5 | 83.3 |
| GLM-5.1 | 83.3 |
| GPT-5.4 Nano | 83.3 |
| GPT-5.5 | 83.3 |
| GPT-5.6 Luna | 83.3 |
| GPT-5.6 Terra | 83.3 |
| Grok 4.20 | 83.3 |
| Kimi K2.6 | 83.3 |
| MiniMax M3 | 83.3 |
| Qwen3.7-Max | 83.3 |
| Gemini 3 Flash Preview | 75.0 |
| GPT-5.4 Mini | 75.0 |
| Grok 4.3 | 75.0 |
| Laguna M.1 | 75.0 |
| MiMo-V2.5-Pro | 75.0 |
| MiniMax M2.7 | 75.0 |
| Muse Spark 1.1 | 75.0 |
| Qwen3.6 Plus (2026-04-02) | 75.0 |
| Qwen3.7-Plus | 75.0 |
| DeepSeek-V4-Pro | 66.7 |
| Gemini 3.1 Flash-Lite Preview | 66.7 |
| Gemini 3.5 Flash-Lite | 66.7 |