| 1 | Qwen3.7 Max | Alibaba | Proprietary | 97.1% | |
| 2 | DeepSeek V4 Pro 0813 | DeepSeek | Proprietary | 95.2% | |
| 3 | DeepSeek V4 Flash 0731 | DeepSeek | Proprietary | 94.8% | |
| 5 | Solar Open 2 | Upstage · Open weight | 93.9% | | |
| 6 | Qwen3.7 Plus | Alibaba · Closed | 92.9% | | |
| 7 | Kimi K2.6 | Moonshot AI · Open weight | 92.7% | | |
| 8 | GLM-5.2 | Z.AI · Open weight | 92.5% | | |
| 10 | Inkling-Small | Thinking Machines Lab · Open weight | 90.2% | | |
| 11 | Qwen3.5 397B | Alibaba · Open weight | 87.9% | | |
| 12 | Qwen3.6 Plus | Alibaba · Closed | 87.8% | | |
| 13 | Kimi K2.5 | Moonshot AI · Open weight | 87.1% | | |
| 14 | Ling 3.0 Flash | InclusionAI · Open weight | 87.0% | | |
| 15 | GLM-5 | Z.AI · Open weight | 86.4% | | |
| 16 | Claude Opus 4.5 | Anthropic · Closed | 85.3% | | |
| 17 | MAI-Thinking-1 | Microsoft · Closed | 84.9% | | |
| 18 | Qwen3.6-27B | Alibaba · Open weight | 84.3% | | |
| 19 | Qwen3.6-35B-A3B | Alibaba · Open weight | 83.6% | | |
| 20 | GLM-5.1 | Z.AI · Open weight | 82.6% | | |
| 21 | K-EXAONE 2.0 | LG AI Research · Open weight | 78.4% | | |
| 22 | ZAYA1-8B | Zyphra · Open weight | 71.6% | | |
| 23 | MiniCPM5-2B | OpenBMB · Open weight | 63.8% | | |
| 25 | LongCat-Flash-Lite-Sparse | Meituan · Open weight | 40.5% | | |
| 27 | MiniCPM5-1B | OpenBMB · Open weight | 25.8% | | |