| 1 | GPT-5.6 Sol | OpenAI | Closed | gpt-5-6-sol | |
| 2 | Kimi K3 | Moonshot AI | Closed | kimi-3 | |
| 3 | Claude Opus 5 | Anthropic | Closed | claude-opus-5 | |
| 4 | GPT-5.5 Pro | OpenAI · Closed | 90.1% | | |
| 5 | GPT-5.4 Pro | OpenAI · Closed | 89.3% | | |
| 6 | Claude Mythos 5 | Anthropic · Closed | 88% | | |
| 7 | GPT-5.6 Terra | OpenAI · Closed | 87.5% | | |
| 8 | Claude Sonnet 5 | Anthropic · Closed | 84.7% | | |
| 9 | GPT-5.5 | OpenAI · Closed | 84.4% | | |
| 10 | Claude Opus 4.8 | Anthropic · Closed | 84.3% | | |
| 11 | Claude Opus 4.6 | Anthropic · Closed | 83.7% | | |
| 12 | MiniMax M3 | MiniMax · Open weight | 83.5% | | |
| 13 | DeepSeek V4 Pro (Max) | DeepSeek · Open weight | 83.4% | | |
| 14 | GPT-5.6 Luna | OpenAI · Closed | 83.3% | | |
| 15 | Kimi K2.6 | Moonshot AI · Open weight | 83.2% | | |
| 16 | GPT-5.4 | OpenAI · Closed | 82.7% | | |
| 17 | DeepSeek V4 Pro (High) | DeepSeek · Open weight | 80.4% | | |
| 18 | Claude Opus 4.7 (Adaptive) | Anthropic · Closed | 79.3% | | |
| 19 | Inkling | Thinking Machines Lab · Open weight | 77.1% | | |
| 20 | Step 3.7 Flash | StepFun · Open weight | 75.8% | | |
| 21 | Agents-A1 | InternScience · Open weight | 75.5% | | |
| 22 | DeepSeek V4 Flash (Max) | DeepSeek · Open weight | 73.2% | | |
| 23 | GLM-5.1 | Z.AI · Open weight | 68% | | |
| 24 | GPT-5.2 | OpenAI · Closed | 65.8% | | |
| 25 | Qwen3.5-122B-A10B | Alibaba · Open weight | 63.8% | | |
| 26 | Qwen3.5 397B | Alibaba · Open weight | 62% | | |
| 27 | Qwen3.5-27B | Alibaba · Open weight | 61% | | |
| 28 | Qwen3.5-35B-A3B | Alibaba · Open weight | 61% | | |
| 29 | Kimi K2.5 | Moonshot AI · Open weight | 60.6% | | |
| 30 | Kimi K2.5 (Reasoning) | Moonshot AI · Closed | 60.6% | | |
| 31 | DeepSeek V4 Flash (High) | DeepSeek · Open weight | 53.5% | | |
| 32 | GLM-4.7 | Z.AI · Open weight | 52% | | |
| 33 | Nemotron 3 Ultra | NVIDIA · Open weight | 44.4% | | |