List your company

Multimodal Video Evaluation on Video-MME (test)

88.6AccuracyGemini 3 Pro
53.55262.65171.7580.849Mar 30, 2026

Evaluation results

MethodLinks
Gemini 3 ProCategory=API call only2026.0388.6
Gemini 2.5 ProCategory=API call only2026.0387.8
Gemini 2.5 FlashCategory=API call only2026.0384.2
GPT-5Category=API call only2026.0383.3
GPT-5 miniCategory=API call only2026.0377.3
Claude Sonnet 4.5Category=API call only2026.0374.2
Keye-VL-1.5-8BCategory=Open weights2026.0373
Eagle2.5-8BCategory=Open weights2026.0372.4
Qwen3-VL-8BCategory=Open weights2026.0371.4
Molmo2-8BCategory=Fully open2026.0369.9
Molmo2-4BCategory=Fully open2026.0369.6
MolmoPoint-8BCategory=MolmoPoint2026.0369.6
Qwen3-VL-4BCategory=Open weights2026.0369.3
GLM-4.1V-9BCategory=Open weights2026.0368.2
MiniCPM-V-4.5-8BCategory=Open weights2026.0367.9
InternVL3.5-8BCategory=Open weights2026.0366
InternVL3.5-4BCategory=Open weights2026.0365.4
VideoChat-Flash-7BCategory=Fully open2026.0365.3
Molmo2-O-7BCategory=Fully open2026.0364.9
LLaVA-Video-7BCategory=Fully open2026.0363.3
PLM-8BCategory=Fully open2026.0358.3
PLM-3BCategory=Fully open2026.0354.9