Terminal task completion on Terminal-bench 1.0
51Pass@1Claude Sonnet 4.5
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude Sonnet 4.5OpenSource=false, Agent Framework=Best with Any Agent2026.02 | 51 | |
| LiberCoder-235B-A22BOpenSource=true, Agent Framework=OpenHands2026.02 | 46.1 | |
| Claude Opus 4.1OpenSource=false, Agent Framework=Best with Any Agent2026.02 | 43.8 | |
| Claude Sonnet 4.5OpenSource=false, Agent Framework=OpenHands2026.02 | 42.7 | |
| Minimax-M2OpenSource=true, Agent Framework=Best with Any Agent2026.02 | 42 | |
| Claude Haiku 4.5OpenSource=false, Agent Framework=Best with Any Agent2026.02 | 41.8 | |
| Claude Sonnet 4OpenSource=false, Agent Framework=OpenHands2026.02 | 41.3 | |
| GLM-4.6OpenSource=true, Agent Framework=Best with Any Agent2026.02 | 40.5 | |
| Grok 4OpenSource=false, Agent Framework=Best with Any Agent2026.02 | 39 | |
| Qwen3-Coder-480B-A35B-InstructOpenSource=true, Agent Framework=Best with Any Agent2026.02 | 39 | |
| LiberCoder-32BOpenSource=true, Agent Framework=OpenHands2026.02 | 38.9 | |
| Qwen3-Coder-30B-A3B-InstructOpenSource=true, Agent Framework=Best with Any Agent2026.02 | 31.3 | |
| Kimi-K2-InstructOpenSource=true, Agent Framework=Best with Any Agent2026.02 | 30 | |
| Qwen3-Coder-30B-A3B-InstructOpenSource=true, Agent Framework=OpenHands2026.02 | 26.5 | |
| Gemini 2.5 ProOpenSource=false, Agent Framework=Best with Any Agent2026.02 | 25.3 | |
| Qwen3-235B-A22B-InstructOpenSource=true, Agent Framework=OpenHands2026.02 | 25 | |
| Qwen3-32BOpenSource=true, Agent Framework=OpenHands2026.02 | 10.3 |