Terminal-based agent interaction on Terminal-Bench 2.0
59.3Success Rate (SR)Claude-Opus-4.5
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Claude-Opus-4.52026.06 | 59.3 | — | |
| Gemini-3-Pro2026.06 | 54.2 | — | |
| GPT-5.12026.06 | 47.6 | — | |
| DeepSeek-V3.22026.06 | 46.4 | — | |
| GLM-4.72026.06 | 41 | — | |
| HarnessBridgeModel=Qwen3.5-35B-A3B, Harness Category=Auto-Harness2026.06 | 33.7 | 1.23 | |
| Meta-HarnessModel=Qwen3.5-35B-A3B, Harness Category=Auto-Harness2026.06 | 31.5 | 2.2 | |
| Terminus 2Model=Qwen3.5-35B-A3B, Harness Category=Manual Harness2026.06 | 30.3 | 2.31 | |
| mini-SWE-agentModel=Qwen3.5-35B-A3B, Harness Category=Manual Harness2026.06 | 29.2 | 6.32 | |
| Kimi-K2-Instruct-09052026.06 | 27.8 | — | |
| Terminus-KIRAModel=Qwen3.5-35B-A3B, Harness Category=Manual Harness2026.06 | 27 | 9.59 | |
| OpenHandsModel=Qwen3.5-35B-A3B, Harness Category=Manual Harness2026.06 | 27 | 2.61 | |
| Qwen-CoderModel=Qwen3.5-35B-A3B, Harness Category=Manual Harness2026.06 | 24.7 | 4.19 | |
| Qwen3-Coder-480B-A35B-Instruct2026.06 | 23.6 | — | |
| LoopCoder-v2R (number of refinement loops)=22026.06 | 21 | — | |
| HarnessBridgeModel=GLM-4.7-Flash, Harness Category=Auto-Harness2026.06 | 20.2 | 0.42 | |
| Terminus 2Model=GLM-4.7-Flash, Harness Category=Manual Harness2026.06 | 19.1 | 1.87 | |
| Meta-HarnessModel=GLM-4.7-Flash, Harness Category=Auto-Harness2026.06 | 14.6 | 1.33 | |
| OpenHandsModel=GLM-4.7-Flash, Harness Category=Manual Harness2026.06 | 13.5 | 1.28 | |
| Qwen3-235B-A22B-Instruct-25072026.06 | 13.5 | — | |
| LoopCoder-v2R (number of refinement loops)=12.22026.06 | 12.2 | — | |
| mini-SWE-agentModel=GLM-4.7-Flash, Harness Category=Manual Harness2026.06 | 11.2 | 1.6 | |
| LoopCoder-v2R (number of refinement loops)=12026.06 | 11.2 | — | |
| Qwen-CoderModel=GLM-4.7-Flash, Harness Category=Manual Harness2026.06 | 10.1 | 1.41 | |
| LoopCoder-v2R (number of refinement loops)=42026.06 | 9 | — | |
| Terminus-KIRAModel=GLM-4.7-Flash, Harness Category=Manual Harness2026.06 | 6.7 | 4.9 | |
| Qwen2.5-Coder-32B-Instruct2026.06 | 4.5 | — | |
| Seed-Coder-8B-Instruct2026.06 | 2.5 | — | |
| Kimi-Dev-72B2026.06 | 2.3 | — | |
| DeepSeek-Coder-V2-Lite-Instruct2026.06 | 0 | — | |
| Qwen2.5-Coder-7B-Instruct2026.06 | 0 | — | |
| Qwen2.5-Coder-14B-Instruct2026.06 | 0 | — |