List your company

Egocentric Video Understanding on EgoSchema (test)

75.6AccuracyGPT-5
18.19233.0964862.904Mar 28, 2026Mar 29, 2026Mar 30, 2026

Evaluation results

MethodLinks
GPT-5Category=API call only2026.0375.6
Claude Sonnet 4.5Category=API call only2026.0373.1
Gemini 2.5 ProCategory=API call only2026.0372.2
Eagle2.5-8BCategory=Open weights2026.0372.2
GPT-5 miniCategory=API call only2026.0370.9
Gemini 2.5 FlashCategory=API call only2026.0370.2
Qwen3-VL-8BCategory=Open weights2026.0369.8
Gemini 3 ProCategory=API call only2026.0368.9
PLM-8BCategory=Fully open2026.0368.8
Qwen3-VL-4BCategory=Open weights2026.0368.4
PLM-3BCategory=Fully open2026.0366.9
MolmoPoint-8BCategory=MolmoPoint2026.0363
GLM-4.1V-9BCategory=Open weights2026.0362.6
Molmo2-8BCategory=Fully open2026.0362
Molmo2-4BCategory=Fully open2026.0361.2
InternVL3.5-4BCategory=Open weights2026.0358.9
InternVL3.5-8BCategory=Open weights2026.0358.6
LLaVA-Video-7BCategory=Fully open2026.0357.3
Molmo2-O-7BCategory=Fully open2026.0356.8
Keye-VL-1.5-8BCategory=Open weights2026.0356.3
VideoChat-Flash-7BCategory=Fully open2026.0351.3
TGPOBackbone=Qwen2.5-VL (3B)2026.0349.7
MiniCPM-V-4.5-8BCategory=Open weights2026.0349.6
EgoVLM GRPOParameters=3B2026.0346.5
LLaVA-VideoParameters=72B2026.0342.7
Llava-OneVisionParameters=72B2026.0336.2
LLaVA-VideoParameters=7B2026.0334.1
Qwen2.5-VLParameters=3B2026.0320.4