OpenVLM MME: leaderboard
OpenCompass OpenVLM evaluation of MME: perception and cognition tests for multimodal models covering OCR, object existence, position, color, celebrity, artwork, commonsense, numerical calculation, and text translation.
Metric: Overall Score. Source: huggingface.co. Status: saturated. 235 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL3-78B | 2538.6 |
| 2 | Qwen 2 VL 72B | 2504.2 |
| 3 | InternVL3-38B | 2500.7 |
| 4 | InternVL2.5-78B | 2467.6 |
| 5 | InternVL3-8B | 2422 |
| 6 | Gemini 2.0 Flash | 2386.1 |
| 7 | GPT-4.1 (2025-04-14) | 2337.6 |
| 8 | Qwen 2 VL 7B | 2276.3 |
| 9 | InternVL2-8B | 2215.1 |
| 10 | Grok 2 (1212) | 2154.7 |
| 11 | Gemini 1.0 Pro | 2148.9 |
| 12 | Gemma 3 27B | 2148.5 |
| 13 | GPT-4.1 Mini | 2110.9 |
| 14 | Gemini 1.5 Pro | 2110.6 |
| 15 | Claude 3.5 Sonnet (20241022) | 2110 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mme · How It Works · Data refreshed daily, snapshot 2026-09-05.