Multimodal Understanding on R1-OneVision 1.0 (test)
59.1Average ScoreGemini-2.0-Flash
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Gemini-2.0-FlashSource Type=Closed-source2026.03 | 59.1 | 56 | 65.9 | 61.2 | 39.8 | 52.3 | 64.4 | 74.3 | 67.2 | 39.8 | |
| Claude-3.5Source Type=Closed-source2026.03 | 52.1 | 56 | 55.9 | 49.4 | 30.6 | 46.5 | 54.3 | 66.7 | 65.7 | 30.6 | |
| Qwen2.5-VL-72BSource Type=Open-source2026.03 | 52 | 54.3 | 56.7 | 54.1 | 23.5 | 48.9 | 55.8 | 63.8 | 63.4 | 23.5 | |
| GPT-4oSource Type=Closed-source2026.03 | 49.6 | 51.3 | 56.2 | 45.3 | 26.5 | 41.3 | 52.5 | 71.4 | 63.4 | 26.5 | |
| R1-Onevision-7B-R-TAPSource Type=Open-source2026.03 | 42.4 | 47.4 | 45 | 34.9 | 31.4 | 39.4 | 36.9 | 55.9 | 59.6 | 31.2 | |
| R1-Onevision-7BSource Type=Open-source2026.03 | 36.2 | 40.1 | 39.5 | 27.6 | 26.5 | 33 | 30.2 | 49.5 | 53 | 26.5 | |
| InternVL2.5-8B-MPOSource Type=Open-source2026.03 | 32.5 | 37.4 | 33.6 | 24.7 | 26.5 | 28.7 | 29.9 | 41 | 44.8 | 26.5 | |
| Qwen2.5-VL-7BSource Type=Open-source2026.03 | 32.1 | 35.3 | 37.1 | 25.9 | 19.4 | 31.5 | 27.3 | 39 | 47 | 19.4 | |
| MiniCPM-o-2.6Source Type=Open-source2026.03 | 30.4 | 33.4 | 31.7 | 21.2 | 31.6 | 24.2 | 31.7 | 30.5 | 41.8 | 31.6 | |
| Qwen2-VL-7BSource Type=Open-source2026.03 | 30 | 35.4 | 32.5 | 25.9 | 26.5 | 26.3 | 28.1 | 30.5 | 45.5 | 26.5 | |
| DeepSeek-VL2Source Type=Open-source2026.03 | 29.8 | 34.4 | 30.9 | 18.8 | 30.6 | 23.5 | 28.4 | 29.5 | 47.8 | 30.6 | |
| InternVL2.5-8BSource Type=Open-source2026.03 | 29.5 | 33.1 | 30.6 | 21.8 | 27.6 | 26.3 | 24.8 | 32.4 | 46.3 | 27.6 |