Share your thoughts, 1 month free Claude Pro on usSee more

Video Question Answering on Molmo QA Benchmarks Long Video 19

80.4Long Video Average

Gemini 2.5 Pro

Updated 19d ago

Evaluation Results

Method	Links
Gemini 2.5 Pro 2026.03		80.4
Gemini 3 Pro 2026.03		78.8
GPT-5 2026.03		76.3
Gemini 2.5 Flash 2026.03		74.5
GPT-5 mini 2026.03		69.8
Claude Sonnet 4.5 2026.03		66.4
Eagle2.5-8B 2026.03		65.2
Molmo2-4B 2026.03		64.5
MolmoPoint-8B 2026.03		64.2
Molmo2-8B 2026.03		64.1
Qwen3-VL-8B 2026.03		63.5
Qwen3-VL-4B 2026.03		62.7
GLM-4.1V-9B 2026.03		60.5
Keye-VL-1.5-8B 2026.03		60.4
MiniCPM-V-4.5-8B 2026.03		60.1
Molmo2-O-7B 2026.03		59.2
VideoChat-Flash-7B 2026.03		58.1
InternVL3.5-8B 2026.03		57.1
InternVL3.5-4B 2026.03		56.5
PLM-8B 2026.03		56.2
LLaVA-Video-7B 2026.03		56.2
PLM-3B 2026.03		53.5