List your company
ResearchFormerly WizWand

Visual Question Answering on MMMU-Pro

49.27AccuracyV3Fusion-Rectify
1.357213.796126.23538.6739Feb 15, 2026Mar 2, 2026Mar 18, 2026Apr 3, 2026Apr 19, 2026May 5, 2026May 21, 2026
Updated 5mo ago

Evaluation results

MethodLinks
V3Fusion-RectifyModel ID=2342026.0349.27
V3Fusion-MLPModel ID=2342026.0347.34
Qwen2.5-VL-7b-InstructModel ID=52026.0346.98
V3Fusion-LEDModel ID=2342026.0346.26
Intern-VL2-8bModel ID=62026.0343.09
Vision-SR1-7BTraining Data=56K2026.0541.1
Qwen2.5-VL-7B-Instruct + Faithful-MR1Backbone model=Qwen2.5-VL-7B-Instruct, Training Data=19.2K2026.0539.7
DeepSeek-VL2-SmallModel ID=42026.0338
Qwen2.5-VL-7B-Instruct + VPPOBackbone model=Qwen2.5-VL-7B-Instruct, Training Data=19.2K2026.0537.8
Qwen2.5-VL-7B-Instruct + GRPOBackbone model=Qwen2.5-VL-7B-Instruct, Training Data=19.2K2026.0536.6
Perception-R1-7BTraining Data=1.4K2026.0535.4
Vision-R1-7BTraining Data=210K2026.0534.7
DeepSeek-VL2-TinyModel ID=32026.0334.61
Qwen2.5-VL-7B-InstructBackbone model=Qwen2.5-VL-7B-Instruct2026.0534.3
LlaVA-v1.6-Vicuna-13bModel ID=12026.0333.69
LlaVA-v1.6-Vicuna-7bModel ID=22026.0332.81
LaViDa-R1Model Category=Unified-Understanding-and-Generation dLLMs2026.0232.8
LaViDa-O +SFTModel Category=Unified-Understanding-and-Generation dLLMs2026.0231.9
Qwen2.5-VL-3B-Instruct + Faithful-MR1Backbone model=Qwen2.5-VL-3B-Instruct, Training Data=19.2K2026.0531.7
LaViDa-OModel Category=Unified-Understanding-and-Generation dLLMs2026.0231.2
Qwen2.5-VL-3B-Instruct + GRPOBackbone model=Qwen2.5-VL-3B-Instruct, Training Data=19.2K2026.0530.2
Qwen2.5-VL-3B-Instruct + VPPOBackbone model=Qwen2.5-VL-3B-Instruct, Training Data=19.2K2026.0529.8
Qwen2.5-VL-3B-InstructBackbone model=Qwen2.5-VL-3B-Instruct2026.0528
LaViDa-LModel Category=Visual-Understanding-Only dLLMs2026.0227.1
MMaDa-8B-Base +CoT SFTModel Category=Unified-Understanding-and-Generation dLLMs2026.028.4
MMaDa-8B-BaseModel Category=Unified-Understanding-and-Generation dLLMs2026.023.2