| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.6 | Anthropic | Closed | claude-opus-4-6 | |
| 2 | Claude Sonnet 4.6 | Anthropic | Closed | claude-sonnet-4-6 | |
| 3 | Qwen 3.6 Max (preview) | Alibaba | Closed | qwen3-6-max-preview | |
| 4 | Qwen3.7 Max | Alibaba · Closed | 73.6% | ||
| 5 | Qwen3.6 Plus | Alibaba · Closed | 71.6% | ||
| 6 | Qwen3.7 Plus | Alibaba · Closed | 71.4% | ||
| 7 | Claude Opus 4.5 | Anthropic · Closed | 70.6% | ||
| 8 | Qwen3.5 397B | Alibaba · Open weight | 70.4% | ||
| 9 | Kimi K2.5 | Moonshot AI · Open weight | 69.2% | ||
| 10 | Qwen3.5-122B-A10B | Alibaba · Open weight | 67.1% | ||
| 11 | GLM-5 | Z.AI · Open weight | 66.8% | ||
| 12 | Qwen3.6-27B | Alibaba · Open weight | 66% | ||
| 13 | Qwen3.5-27B | Alibaba · Open weight | 65.6% | ||
| 14 | Qwen3.6-35B-A3B | Alibaba · Open weight | 64.7% | ||
| 15 | Qwen3.5-35B-A3B | Alibaba · Open weight | 63.4% | ||
| 16 | Qwen3 235B 2507 | Alibaba · Open weight | 62.6% | ||
| 17 | DeepSeek V4 Pro Base | DeepSeek · Open weight | 53.9% | ||
| 18 | DeepSeek V4 Flash Base | DeepSeek · Open weight | 46.5% | ||
| 19 | MiniCPM5-1B | OpenBMB · Open weight | 23.1% |