GMMLU (Anthropic): leaderboard
Anthropic system-card run of Global MMLU, measuring multilingual academic and professional knowledge.
Metric: Average Accuracy (%). Source: www-cdn.anthropic.com. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 92.2 | #58 |
| 2 | GPT-5.4 | 90.6 | #79 |
| 3 | Claude Opus 4.8 | 90.4 | #37 |
| 4 | Claude Opus 4.7 | 89.9 | #49 |
| 5 | Claude Sonnet 4.6 | 86.1 | #90 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=gmmlu-anthropic · How It Works · Data refreshed daily, snapshot 2026-10-20.