| 1 | GPT-6 Astra | OpenAI | Proprietary | 95% | |
| 2 | GPT-5.6 Sol | OpenAI | Proprietary | 92.5% | |
| 3 | Claude Opus 5 | Anthropic | Proprietary | 90.4% | |
| 4 | Claude Fable 5.1 | Anthropic · Closed | 90% | | |
| 5 | GPT-5.5 | OpenAI · Closed | 85% | | |
| 6 | GPT-5.6 Terra | OpenAI · Closed | 83.9% | | |
| 7 | GPT-5.4 Pro | OpenAI · Closed | 83.3% | | |
| 8 | dots3-note Preview | Dots Studio · Open weight | 81.4% | | |
| 9 | Gemini 3.1 Pro | Google · Closed | 77.1% | | |
| 10 | Claude Opus 4.7 (Adaptive) | Anthropic · Closed | 75.8% | | |
| 11 | GPT-5.4 | OpenAI · Closed | 74.0% | | |
| 12 | Gemini 3.5 Flash | Google · Closed | 72.1% | | |
| 13 | Claude Opus 4.8 | Anthropic · Closed | 72.1% | | |
| 14 | GPT-5.6 Luna | OpenAI · Closed | 59.5% | | |
| 15 | GPT-5.2 Pro | OpenAI · Closed | 54.2% | | |
| 16 | Grok 4.20 | xAI · Closed | 53.3% | | |
| 17 | GPT-5.2 | OpenAI · Closed | 52.9% | | |
| 18 | Grok 4.5 | xAI · Closed | 52.6% | | |
| 19 | Gemini 3 Pro Deep Think | Google · Closed | 45.1% | | |
| 20 | Muse Spark | Meta · Closed | 42.5% | | |
| 21 | Inkling-Small | Thinking Machines Lab · Open weight | 40.1% | | |
| 22 | Gemini 3 Pro | Google · Closed | 31.1% | | |
| 23 | Claude Sonnet 4.5 | Anthropic · Closed | 13.6% | | |