List your company

Image Quality Understanding on Q-Bench subset (dev)

85.82Yes/No AccuracyQualiRAG
62.586468.618274.6580.6818Jan 26, 2026

Evaluation results

MethodLinks
QualiRAGRequires VQA-labeled data=false2026.0185.8281.7484.674.5977.0181.7185.287.35
VisualQuality-R1Requires VQA-labeled data=true2026.0182.917982.371.677.6377.7879.6183.27
Gemini-3-ProRequires VQA-labeled data=false2026.0182.9180.6388.4871.0577.3378.8383.886.67
GPT-5Requires VQA-labeled data=false2026.0182.1880.2784.5174.2477.6379.8680.9285.71
Co-InstructRequires VQA-labeled data=true2026.018275.3876.7766.7375.2971.5374.6783.27
Q-InsightRequires VQA-labeled data=true2026.018277.6682.368.5676.0776.3977.383.67
Qwen2.5-VL-7BRequires VQA-labeled data=false2026.0181.4577.8683.1968.9775.176.8577.9685.31
GLM-4.1V-9BRequires VQA-labeled data=false2026.0181.4577.3980.5369.9876.0775.2376.6484.9
LLaVA-OneVision-7BRequires VQA-labeled data=false2026.0181.2776.7980.3168.5673.7477.7874.6784.08
Qwen3-VL-8BRequires VQA-labeled data=false2026.0180.9178.5383.8570.9977.4377.0878.6283.27
InternVL3.5-8BRequires VQA-labeled data=false2026.0180.3676.7279.8769.7875.174.5476.6484.08
InternVL3-9BRequires VQA-labeled data=false2026.0180.1875.9978.7668.7672.7675.6973.6886.12
Q-InstructRequires VQA-labeled data=true2026.0176.9166.3565.0455.7864.0167.1364.871.84
mPLUG-Owl3-7BRequires VQA-labeled data=false2026.0163.4863.477.8550.8757.866.8967.3364