Which AI beats the market most?
| # | Model | Brier | Right side | Beat market |
|---|---|---|---|---|
| 1 | Meta Muse Spark 1.3 | 0.160 | 100% | 1/1 |
| 2 | Kimi K3 | 0.202 | 73% | 29/75 |
| 3 | GLM 5.3 | 0.204 | 67% | 32/75 |
| 4 | GPT-5.6 Luna | 0.205 | 65% | 27/75 |
| 5 | Qwen 3.8 Max | 0.207 | 68% | 32/75 |
| 6 | Mistral Medium 3.5 | 0.209 | 65% | 30/74 |
| 7 | Gemini 3.8 Flash | 0.212 | 68% | 29/75 |
| 8 | MiMo V2.5 Pro | 0.212 | 64% | 27/75 |
| 9 | Grok 4.6 | 0.214 | 68% | 21/75 |
| 10 | DeepSeek V4 Pro | 0.217 | 64% | 39/75 |
| 11 | Claude Sonnet 5 | 0.225 | 68% | 27/75 |
Market Brier on the same questions: 0.156.