OpenVLM MMBench V1.1 CN - Object Localization: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 90.5 |
| 2 | InternVL3-78B | 90.5 |
| 3 | InternVL3-38B | 90.5 |
| 4 | InternVL2.5-78B | 88.6 |
| 5 | Qwen 2 VL 72B | 86.7 |
| 6 | GPT-4.5 | 85.7 |
| 7 | GPT-5 Mini (2025-08-07) | 84.8 |
| 8 | GPT-5 | 83.8 |
| 9 | GPT-4o ChatGPT | 82.9 |
| 10 | Grok 2 (1212) | 81.9 |
| 11 | GPT-4.1 (2025-04-14) | 81.9 |
| 12 | GPT-4.1 Mini | 79 |
| 13 | InternVL3-8B | 79 |
| 14 | Claude 3.7 Sonnet | 79 |
| 15 | Claude 3.5 Sonnet | 78.1 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmbench-v1-1-cn-object-localization · How It Works · Data refreshed daily, snapshot 2026-09-05.