LLM Stats (MMMLU): leaderboard
Original MMLU translated into major world languages (Chinese, French, Arabic, Hindi, etc.), measuring knowledge transfer across languages.
Metric: Score (%). Source: llm-stats.com. Source: qwen.ai. Saturation forecast: Estimated already saturated. 51 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Mythos Preview | 92.7 | #4 |
| 2 | Gemini 3.1 Pro (Preview) | 92.6 | #58 |
| 3 | Gemini 3 Flash | 91.8 | #99 |
| 4 | Gemini 3 Pro | 91.8 | #80 |
| 5 | Claude Opus 4.7 | 91.5 | #49 |
| 6 | Claude Opus 4.6 | 91.1 | #63 |
| 7 | Claude Opus 4.5 | 90.8 | #83 |
| 8 | Claude Opus 4.6 (Thinking) | 90.6 | #63 (Claude Opus 4.6) |
| 9 | Qwen 3.7 Max | 90.3 | #75 |
| 10 | GPT-5.2 | 89.6 | #110 |
| 11 | Qwen 3.6 Plus | 89.5 | #120 |
| 12 | Claude Opus 4.1 | 89.5 | #140 |
| 13 | Claude Sonnet 4.6 | 89.3 | #90 |
| 14 | Claude Sonnet 4.5 | 89.1 | #142 |
| 15 | Qwen 3.7 Plus | 89 | #92 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-mmmlu · How It Works · Data refreshed daily, snapshot 2026-10-20.