MMMU-Pro No Tools (Opus 4.6 System Card): leaderboard

MMMU-Pro No Tools results as reported in the Claude Opus 4.6 system card — Anthropic-run evaluation comparing a small set of frontier models.

Metric: Score (%). Source: www-cdn.anthropic.com. Saturation forecast: Around January 2027. 5 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro81#80
2GPT-5.279.5#110
3Claude Opus 4.6 (Thinking)73.9#63 (Claude Opus 4.6)
4Claude Opus 4.5 (Thinking)70.6#83 (Claude Opus 4.5)
5Claude Sonnet 4.563.4#142

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=mmmu-pro-no-tools-opus-4-6-system-card · How It Works · Data refreshed daily, snapshot 2026-10-20.