Coding on SWE-bench Multilingual
77.5ScoreClaude Opus 4.5
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude Opus 4.5Framework=OpenHands2026.02 | 77.5 | |
| GLM-5Framework=OpenHands2026.02 | 73.3 | |
| Kimi K2.5Framework=OpenHands2026.02 | 73 | |
| GPT-5.2 (xhigh)Framework=OpenHands2026.02 | 72 | |
| Gemini-3-Flash High2026.06 | 71.1 | |
| DeepSeek-V3.2Framework=OpenHands2026.02 | 70.2 | |
| GLM-4.7Framework=OpenHands2026.02 | 66.7 | |
| Gemini 3 ProFramework=OpenHands2026.02 | 65 | |
| Seed2.0 Lite2026.06 | 64.4 | |
| GPT-5-mini High2026.06 | 63.8 |