MMBench: leaderboard
MMBench: Evaluates multimodal understanding across image, text, chart, diagram, or cross-modal reasoning tasks.
Metric: MMBench-EN (%). Source: opencompass.org.cn. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 9B | 90.1 | #379 |
| 2 | Qwen 3.5 4B | 89.4 | #475 |
| 3 | InternVL3-78B | 89 | #352 |
| 4 | GPT-5 Nano | 80.3 | #425 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mmbench · How It Works · Data refreshed daily, snapshot 2026-10-20.