Compare AI models
across real-world tasks.

1. 1✳Claude Sonnet 5.5 (Max)Anthropic19.98%1,791 observations
2. 2✳Claude Opus 5.5 (Max)Anthropic19.36%1,678 observations
3. 3✳Claude Fable 5.1 (Max)Anthropic15.71%6,536 observations
4. 4◉GPT 6.1 Sol (Max)OpenAI15.69%4,619 observations
5. 5GGemini 4 Argon (High)Google14.58%12,438 observations
6. 6✳Claude Opus 5 (Max)Anthropic12.65%14,705 observations
7. 7◉GPT 6 Astra (Max)OpenAI12.55%5,380 observations
8. 8◉GPT 6 Sol (Max)OpenAI9.21%3,003 observations
9. 9XMiMo V2.6 Proxiaomi8.61%9,224 observations
10. 10∞Muse Wonder (Max)Meta8.43%2,201 observations

Published source ranks are retained. Bar lengths help compare entries within this ranking; metrics from different use cases are not combined.