OpenVLM MMT-Bench - LVLM Response Judgement: leaderboard
The "LVLM Response Judgement" subtask of OpenVLM MMT-Bench. OpenCompass OpenVLM evaluation of MMT-Bench: multimodal tasks covering visual recognition, localization, OCR, counting, relation reasoning, visual math, and hallucination detection.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 207 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 (2025-04-14) | 70 | #250 |
| 2 | InternVL3-78B | 65 | #352 |
| 3 | InternVL2-8B | 65 | #817 |
| 4 | GPT-4.1 Nano | 60 | #731 |
| 5 | InternVL3-14B | 60 | #505 |
| 6 | Qwen 2 VL 72B | 60 | #460 |
| 7 | Gemini 1.5 Pro | 60 | #445 |
| 8 | Gemma 3 4B | 60 | #1089 |
| 9 | Grok 2 (1212) | 60 | #496 |
| 10 | Llama 3.2 90B Vision Instruct | 60 | #622 |
| 11 | GPT-4.1 Mini | 55 | #355 |
| 12 | MiniCPM-V-2.6 | 55 | #836 |
| 13 | Pixtral-12B | 55 | #808 |
| 14 | Aquila-VL-2B | 55 | #1238 |
| 15 | Gemma 3 12B | 55 | #671 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmt-bench-lvlm-response-judgement · How It Works · Data refreshed daily, snapshot 2026-10-20.