| 1 | Qwen3.8 Max | Alibaba | Open weights | 93.5% | |
| 2 | Claude Mythos 5 | Anthropic | Proprietary | 93.5% | |
| 3 | Qwen3.8-Omni-Flash | Alibaba | Proprietary | 91.4% | |
| 4 | Kimi K3 | Moonshot AI · Closed | 91.3% | | |
| 5 | Claude Opus 4.7 (Adaptive) | Anthropic · Closed | 91% | | |
| 6 | Qwen3.8-Flash-Next | Alibaba · Open weight | 90.6% | | |
| 7 | Qwen3.8-27B | Alibaba · Open weight | 90.2% | | |
| 8 | Claude Opus 4.8 | Anthropic · Closed | 89.9% | | |
| 9 | GLM-5.3-Flash | Z.AI · Open weight | 89.4% | | |
| 10 | Gemini 3.7 Flash | Google · Closed | 88.7% | | |
| 11 | Muse Spark 1.1 | Meta · Closed | 88.4% | | |
| 12 | Claude Sonnet 5 | Anthropic · Closed | 88.3% | | |
| 13 | Sakana Fugu-Ultra | Sakana AI · Closed | 86.6% | | |
| 14 | Muse Spark | Meta · Closed | 86.4% | | |
| 15 | Qwen3.7 Plus | Alibaba · Closed | 85.9% | | |
| 16 | Seed 2.1 Pro | ByteDance · Closed | 85.4% | | |
| 17 | Sakana Fugu | Sakana AI · Closed | 85.1% | | |
| 18 | Gemini 3.5 Flash | Google · Closed | 84.2% | | |
| 19 | GPT-5.4 | OpenAI · Closed | 82.8% | | |
| 20 | Seed 2.1 Turbo | ByteDance · Closed | 82.5% | | |
| 21 | GPT-5.2 | OpenAI · Closed | 82.1% | | |
| 22 | Inkling | Thinking Machines Lab · Open weight | 82% | | |
| 23 | Qwen3.6 Plus | Alibaba · Closed | 81.5% | | |
| 24 | Gemini 3 Pro | Google · Closed | 81.4% | | |
| 25 | Inkling-Small | Thinking Machines Lab · Open weight | 81.3% | | |
| 26 | MiMo-V2.5 | Xiaomi · Closed | 81% | | |
| 27 | Qwen3.5 397B | Alibaba · Open weight | 80.8% | | |
| 28 | Kimi K2.6 | Moonshot AI · Open weight | 80.4% | | |
| 29 | Gemini 3.1 Pro | Google · Closed | 80.2% | | |
| 30 | Muse Glimmer 30B | Meta · Open weight | 78.8% | | |
| 31 | Qwen3.6-27B | Alibaba · Open weight | 78.4% | | |
| 32 | Qwen3.6-35B-A3B | Alibaba · Open weight | 78% | | |
| 33 | Claude Sonnet 4.6 | Anthropic · Closed | 77.4% | | |
| 34 | Qwen3.5-122B-A10B | Alibaba · Open weight | 77.2% | | |
| 35 | Nemotron 3 Nano Omni 30B A3B | NVIDIA · Open weight | 76.3% | | |
| 36 | Gemini 3.1 Flash-Lite | Google · Closed | 73.2% | | |
| 37 | Claude Opus 4.5 | Anthropic · Closed | 68.5% | | |
| 38 | Grok 4.20 | xAI · Closed | 60.9% | | |
| 39 | Command A+ | Cohere · Open weight | 52.7% | | |