| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | Anthropic | Proprietary | 100.00% | |
| 2 | Claude Fable 5.1 | Anthropic | Proprietary | 100.00% | |
| 3 | logicalintelligence | 100.00% | Alephprover | ||
| 4 | GPT-6 Astra | OpenAI · Closed | 99.00% | ||
| 5 | Claude Opus 5 | Anthropic · Closed | 99.00% | ||
| 6 | Claude Fable 5 | Anthropic · Closed | 95.00% | ||
| 7 | Kimi K3 | Moonshot AI · Closed | 87.00% | ||
| 8 | Aristotle | Aristotle | 86.00% | ||
| 9 | GPT-5.6 Sol | OpenAI · Closed | 83.00% | ||
| 10 | Claude Sonnet 5 | Anthropic · Closed | 77.00% | ||
| 11 | Hy4 preview | Tencent · Open weight | 75.00% | ||
| 12 | GPT-5.6 Terra | OpenAI · Closed | 74.00% | ||
| 13 | GPT-5.6 Luna | OpenAI · Closed | 60.00% | ||
| 14 | Gemini 3.7 Flash | Google · Closed | 58.00% | ||
| 15 | Qwen3.8 Max | Alibaba · Open weight | 58.00% | ||
| 16 | DeepSeek V4 Flash 0731 | DeepSeek · Closed | 56.00% | ||
| 17 | DeepSeek V4.1 Flash | DeepSeek · Open weight | 54.00% | ||
| 18 | Grok 4.6 | xAI · Closed | 51.00% | ||
| 19 | DeepSeek V4 Pro 0813 | DeepSeek · Closed | 50.00% | ||
| 20 | GLM-5.3 | Z.AI · Open weight | 49.00% | ||
| 21 | Gemini 3.8 Flash | Google · Closed | 48.00% | ||
| 22 | Muse Spark 1.2 | Meta · Closed | 43.00% | ||
| 23 | Gemini 3.5 Flash | Google · Closed | 31.00% | ||
| 24 | Grok 4.5 | xAI · Closed | 31.00% | ||
| 25 | Grok 4.7 | xAI · Closed | 26.00% | ||
| 26 | Gemini 3.1 Pro Preview | 26.00% | |||
| 27 | MiMo-V2.5-Pro | Xiaomi · Closed | 22.00% | ||
| 28 | GLM-5.3-Flash | Z.AI · Open weight | 21.00% | ||
| 29 | MiniMax M3 | MiniMax · Open weight | 18.00% | ||
| 31 | MiMo-V2.5 | Xiaomi · Closed | 16.00% | ||
| 32 | Qwen3.8-27B | Alibaba · Open weight | 16.00% | ||
| 33 | Mistral Medium 3.5 | Mistral | 9.00% | ||
| 34 | Inkling-Small | Thinking Machines Lab · Open weight | 6.00% | ||
| 35 | Mercury 2.5 | Inception · Closed | 3.00% | ||
| 36 | Inkling | Thinking Machines Lab · Open weight | 0.00% |