List your company
ResearchFormerly WizWand

Multimodal Mathematical Reasoning on MATH-Vision

51.2AccuracyVision-R1 (7B)
20.41628.40836.444.392Aug 27, 2025
Updated 6mo ago

Evaluation results

MethodLinks
Vision-R1 (7B)Backbone=7B2025.0851.244.5
Visionary-R1 (3B)Backbone=3B2025.0840.433.9
Zero-shot InferenceBackbone=Qwen2.5-VL-3B, Inference Protocol=Zero-shot (before RL)2025.0840.435.5
Vision-SR1Backbone=Qwen2.5-VL-7B2025.0836.252.2
Percention-R1 (7B)Backbone=7B2025.0835.745.2
Vision-R1Backbone=Qwen2.5-VL-7B, Training Data=47K data2025.0833.850.7
Zero-shot InferenceBackbone=Qwen2.5-VL-7B, Inference Protocol=Zero-shot (before RL)2025.0831.941.5
Vision-R1Backbone=Qwen2.5-VL-3B, Training Data=47K data2025.0829.947.1
Vision-SR1Backbone=Mimo-VL-7B2025.0829.649.5
Vision-SR1Backbone=Qwen2.5-VL-3B2025.0829.348.8
Vision-R1Backbone=Mimo-VL-7B, Training Data=47K data2025.0825.746
Zero-shot InferenceBackbone=Mimo-VL-7B, Inference Protocol=Zero-shot (before RL)2025.0821.644.4