OpenVLM MMBench V1.1 EN - Function Reasoning: leaderboard
OpenCompass OpenVLM MMBench V1.1 English split accuracy for function reasoning.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 284 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL2.5-78B | 98.9 | #431 |
| 2 | Llama 3.2 90B Vision Instruct | 97.8 | #622 |
| 3 | Gemini 1.5 Flash (002) | 96.7 | #583 |
| 4 | Gemini 1.5 Flash | 96.7 | #618 |
| 5 | Grok 2 (1212) | 96.7 | #496 |
| 6 | InternVL3-78B | 95.6 | #352 |
| 7 | GPT-5 Mini (2025-08-07) | 95.6 | #172 |
| 8 | Gemma 3 12B | 95.6 | #671 |
| 9 | Qwen 2.5 VL 7B | 95.6 | #643 |
| 10 | Gemini 1.5 Pro (002) | 94.4 | #349 |
| 11 | InternVL3-38B | 94.4 | #405 |
| 12 | GPT-4.1 (2025-04-14) | 94.4 | #250 |
| 13 | MiniCPM-V-2.6 | 94.4 | #836 |
| 14 | Qwen 2 VL 72B | 94.4 | #460 |
| 15 | Pixtral-12B | 94.4 | #808 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmbench-v1-1-en-function-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-20.