OpenVLM MMT-Bench - Artwork Emotion Recognition: leaderboard
The "Artwork Emotion Recognition" 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. 207 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2 VL 2B | 70 | #1220 |
| 2 | Qwen 2 VL 7B | 65 | #750 |
| 3 | GPT-4.1 Nano | 55 | #731 |
| 4 | InternVL3-8B | 55 | #620 |
| 5 | GPT-4.1 (2025-04-14) | 55 | #250 |
| 6 | MiniCPM-V-2.6 | 55 | #836 |
| 7 | Gemini 1.5 Pro | 55 | #445 |
| 8 | Aquila-VL-2B | 55 | #1238 |
| 9 | Qwen 2.5 VL 7B | 55 | #643 |
| 10 | Llama 3.2 11B Instruct | 55 | #1126 |
| 11 | Gemini 1.5 Flash | 55 | #618 |
| 12 | InternVL3-78B | 50 | #352 |
| 13 | Gemma 3 4B | 50 | #1089 |
| 14 | GPT-4.1 Mini | 45 | #355 |
| 15 | InternVL3-38B | 45 | #405 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmt-bench-artwork-emotion-recognition · How It Works · Data refreshed daily, snapshot 2026-10-20.