Reasoning Step Judgment on VisualProcessBench
74.1MMMU ScoreAthena-PRM
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Athena-PRMEvaluation Protocol=Multimodal PRMs, Model Size (Parameters)=7B2025.06 | 74.1 | 61.2 | 60.7 | 72.7 | 73.8 | 65.9 | |
| Qwen2.5-VL-72BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=72B2025.06 | 59.2 | 59 | 59.7 | 62.9 | 62.3 | 60.5 | |
| Gemini-2.0-FlashEvaluation Protocol=Proprietary Models as Judge2025.06 | 58.5 | 60.1 | 62.8 | 66.7 | 58.7 | 62.3 | |
| VisualPRM-8BEvaluation Protocol=Multimodal PRMs, Model Size (Parameters)=8B2025.06 | 58.5 | 62.1 | 61 | 62.7 | 61.8 | 62 | |
| GPT-4oEvaluation Protocol=Proprietary Models as Judge2025.06 | 56.3 | 60.2 | 59.7 | 59 | 63.3 | 60.3 | |
| GPT-4o-miniEvaluation Protocol=Proprietary Models as Judge2025.06 | 53.6 | 58.9 | 57.1 | 56.7 | 58.5 | 57.9 | |
| Qwen2.5-VL-7BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=7B2025.06 | 53.1 | 51.8 | 47.8 | 51.3 | 54.2 | 51 | |
| InternVL2.5-78BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=78B2025.06 | 52 | 51.7 | 53.7 | 50.8 | 52.5 | 52.6 | |
| InternVL2.5-38BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=38B2025.06 | 51.5 | 48.4 | 50.9 | 51.8 | 52.5 | 50.8 | |
| Random GuessingModel Category=Baseline2025.06 | 50 | 50 | 50 | 50 | 50 | 50 | |
| InternVL2.5-26BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=26B2025.06 | 48.8 | 47.4 | 49.2 | 50.4 | 51.4 | 49.2 | |
| InternVL2.5-8BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=8B2025.06 | 47.1 | 45.5 | 47.8 | 50.3 | 50.8 | 48 | |
| LLaVA-OV-72BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=72B2025.06 | 46.1 | 48.4 | 53 | 57 | 57.3 | 52.3 | |
| LLaVA-OV-7BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=7B2025.06 | 45.7 | 43 | 42.2 | 44.7 | 52.5 | 44.4 | |
| MiniCPM-V2.6-8BEvaluation Protocol=Open-source Models as Judge, Model Size (Parameters)=8B2025.06 | 44.9 | 50.9 | 58.9 | 46.7 | 57.4 | 50.4 |