Vision-Centric Reasoning on MuirBench
68AccuracyGPT-4o
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4omodel_category=Proprietary Model2025.09 | 68 | |
| Gemini2.0-Flashmodel_category=Proprietary Model2025.09 | 64.6 | |
| MM-Eureka-Qwen-7Bbase_model=Qwen2.5-VL-7B2025.09 | 61.1 | |
| VisionZero-Qwen-7B (Real-World)training_context=Real-World, base_model=Qwen2.5-VL-7B2025.09 | 59.4 | |
| VisionZero-Qwen-7B (Chart)training_context=Chart, base_model=Qwen2.5-VL-7B2025.09 | 59.2 | |
| Qwen2.5-VL-7Bbase_model=Qwen2.5-VL-7B2025.09 | 58.2 | |
| VisionZero-Qwen-7B (CLEVR)training_context=CLEVR, base_model=Qwen2.5-VL-7B2025.09 | 58.2 | |
| ViGaL-Snake+Rotationbase_model=Qwen2.5-VL-7B2025.09 | 57.8 | |
| VLAA-Thinker-7Bbase_model=Qwen2.5-VL-7B2025.09 | 57.1 | |
| OpenVLThinker-7Bbase_model=Qwen2.5-VL-7B2025.09 | 52.8 | |
| R1-OneVision-7Bbase_model=Qwen2.5-VL-7B2025.09 | 46.3 |