List your company

Visual Question Answering on In-Domain Multiple-Choice

92.9OCR-VQA AccuracyEXTRACT+THINK
39.02853.0146780.986Nov 21, 2025

Evaluation results

MethodLinks
EXTRACT+THINKPerception Module (P)=1.7B, Reasoning Module (R)=4.0B, #Vis. Data=2.4M2025.1192.990.175.284.477.891.385.3
PrismCaptionerPerception Module (P)=7.0B, Reasoning Module (R)=70B, #Vis. Data=1.9M2025.1191.57768.179.967.585.878.3
EXTRACT+THINK†Perception Module (P)=1.7B, Reasoning Module (R)=4.0B, #Vis. Data=0.4M2025.1191.58471.384.677.886.982.7
EXTRACT+THINKPerception Module (P)=0.6B, Reasoning Module (R)=1.7B, #Vis. Data=2.4M2025.1189.481.872.27874.785.680.3
PrismCaptionerPerception Module (P)=1.8B, Reasoning Module (R)=70B, #Vis. Data=1.9M2025.1189.272.764.677.86682.375.4
CAPTION+THINKPerception Module (P)=1.7B, Reasoning Module (R)=4.0B, #Vis. Data=2.0M2025.1189.284.868.980.572.184.380
EXTRACT+THINK†Perception Module (P)=0.6B, Reasoning Module (R)=1.7B, #Vis. Data=0.4M2025.1186.979.869.976.672.582.178
CAPTION+THINKPerception Module (P)=0.6B, Reasoning Module (R)=1.7B, #Vis. Data=2.0M2025.1184.980.660.674.766.28375
InternVL2.5LLM Size=0.5B, #Vis. Data=64M2025.1179.889.189.88275.48383.2
Baseline (from §3.3)Perception Module (P)=1.7B, Reasoning Module (R)=4.0B, #Vis. Data=1.0M2025.1179.459.46571.664.576.469.4
Baseline (from §3)LLM Size=1.7B, #Vis. Data=1.0M2025.1173.483.476.277.874.375.876.8
SmolVLMLLM Size=1.7B, #Vis. Data=unk.2025.1172.981.479.775.570.675.175.9
Baseline (from §3.3)Perception Module (P)=0.6B, Reasoning Module (R)=4.0B, #Vis. Data=1.0M2025.1171.850.76367.662.372.364.6
LLaVA-OneVisionLLM Size=0.5B, #Vis. Data=8.8M2025.1169.577.255.775.773.674.771.1
Baseline (from §3)LLM Size=0.6B, #Vis. Data=1.0M2025.1141.171.367.971.269.574.565.9