Vals AI MMLU-Pro — leaderboard
Metric: Accuracy (%). Source: www.vals.ai. 122 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 | 91.5 |
| 2 | Gemini 3.1 Pro (Preview) | 90.99 |
| 3 | Gemini 3 Pro (Preview) | 90.1 |
| 4 | Claude Opus 4.7 | 89.87 |
| 5 | Claude Opus 4.8 | 89.58 |
| 6 | Gemini 3.5 Flash | 89.52 |
| 7 | Qwen 3.7 Max | 89.31 |
| 8 | Grok 4.5 | 89.22 |
| 9 | Claude Opus 4.6 (Thinking) | 89.11 |
| 10 | GPT-5.6 Sol | 89.1 |
| 11 | Muse Spark 1.1 | 88.73 |
| 12 | Gemini 3 Flash (Preview) | 88.59 |
| 13 | GPT-5.5 | 88.14 |
| 14 | Kimi K3 | 87.97 |
| 15 | Claude Opus 4.1 (20250805) (Thinking) | 87.92 |
Interactive version: theaggregate.ai/benchmark?slug=vals-ai-mmlu-pro · How the rankings work · Data refreshed daily, snapshot 2026-07-22.