Multilingual Software Engineering on SWE-Bench Multilingual
77.1Success RateKimi-K2.6
Evaluation Results
| Method | Links | |
|---|---|---|
| Kimi-K2.6Parameter Count=1T-A32B2026.06 | 77.1 | |
| DS-v4-ProParameter Count=1.6T-A49B2026.06 | 76.5 | |
| DS-v4-FlashParameter Count=284B-A13B2026.06 | 75 | |
| GLM-5.1Parameter Count=744B-A40B2026.06 | 74.8 | |
| MiniMax-2.7Parameter Count=230B-A10B2026.06 | 71.8 | |
| Qwen-3.5Parameter Count=397B-17B2026.06 | 70.9 | |
| Nemotron 3 UltraParameter Count=550B-A55B2026.06 | 67.7 |