OpenVLM MMMU - History: leaderboard
OpenCompass OpenVLM MMMU validation subcategory score for visual college-level questions in history.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 284 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 86.7 | #152 |
| 2 | GPT-5 | 83.3 | #98 |
| 3 | InternVL3-8B | 83.3 | #620 |
| 4 | Gemini 1.5 Flash (002) | 83.3 | #583 |
| 5 | InternVL3-78B | 83.3 | #352 |
| 6 | InternVL3-14B | 83.3 | #505 |
| 7 | Gemma 3 27B | 83.3 | #607 |
| 8 | Qwen 2 VL 72B | 83.3 | #460 |
| 9 | Gemini 1.5 Flash | 83.3 | #618 |
| 10 | GPT-4.5 | 83.3 | #274 |
| 11 | GPT-4.1 (2025-04-14) | 80 | #250 |
| 12 | GPT-5 Mini (2025-08-07) | 80 | #172 |
| 13 | Claude 3.5 Sonnet | 80 | #350 |
| 14 | Gemma 3 12B | 80 | #671 |
| 15 | InternVL2.5-78B | 80 | #431 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmmu-history · How It Works · Data refreshed daily, snapshot 2026-10-20.