OpenVLM MMBench V1.1 EN - Nature Relation: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 96.7 |
| 2 | InternVL3-78B | 94.6 |
| 3 | GPT-4.1 (2025-04-14) | 94.6 |
| 4 | GPT-4.5 | 94.6 |
| 5 | Grok 2 (1212) | 93.5 |
| 6 | Claude 3.7 Sonnet | 92.4 |
| 7 | GPT-4o ChatGPT | 92.4 |
| 8 | Gemini 1.5 Pro (002) | 91.3 |
| 9 | Gemini 1.5 Flash (002) | 91.3 |
| 10 | InternVL2.5-78B | 91.3 |
| 11 | GPT-5 Mini (2025-08-07) | 91.3 |
| 12 | GPT-4.1 Mini | 90.2 |
| 13 | Gemma 3 27B | 90.2 |
| 14 | Gemini 2.5 Pro | 89.1 |
| 15 | InternVL3-8B | 88 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmbench-v1-1-en-nature-relation · How It Works · Data refreshed daily, snapshot 2026-09-05.