List your company

Visual Question Answering on MuirBench

78.6AccuracyGPT-5
21.29636.17351.0565.927Mar 30, 2026

Evaluation results

MethodLinks
GPT-5Access=API call only2026.0378.6
GLM-4.1V-9BAccess=Open weights only2026.0374.7
Gemini 2.5 ProAccess=API call only2026.0374.5
Gemini 2.5 FlashAccess=API call only2026.0373.5
GPT-5 miniAccess=API call only2026.0371.4
Qwen3-VL-8BAccess=Open weights only2026.0364.4
Qwen3-VL-4BAccess=Open weights only2026.0363.8
Molmo2-8BAccess=Fully open2026.0363.7
MolmoPoint-8BAccess=MolmoPoint2026.0362.5
Eagle2.5-8BAccess=Open weights only2026.0361.8
Molmo2-4BAccess=Fully open2026.0360.5
Claude Sonnet 4.5Access=API call only2026.0359.6
MolmoPoint-8B-O-7BAccess=Fully open2026.0358.4
InternVL3.5-8BAccess=Open weights only2026.0355.8
MiniCPM-V-4.5-8BAccess=Open weights only2026.0353.3
InternVL3.5-4BAccess=Open weights only2026.0353.1
Keye-VL-1.5-8BAccess=Open weights only2026.0351.2
PLM-3BAccess=Fully open2026.0325.7
PLM-8BAccess=Fully open2026.0323.5