List your company

Temporal video understanding on TOMATO

53AccuracyGPT-4o
23.46431.13238.846.468Jun 1, 2026

Evaluation results

MethodLinks
GPT-4o2026.0653
Gemini 1.5 ProSource=[60]2026.0648.6
Gemini 1.5 ProSource=[59]2026.0648.3
GPT-4o mini2026.0644.1
AdaCodecToken budget=comparable, Backbone=Qwen2-VL-7B2026.0640
AdaCodecToken budget=1/7, Backbone=Qwen2-VL-7B2026.0639.8
Claude 3.5 Sonnet2026.0639.6
Molmo2-8B2026.0639.6
Gemini 1.5 Flash2026.0639.1
Molmo2-O-7B2026.0636.2
Qwen2-VL-7B2026.0635.7
PLM-7B2026.0633.2
Key-VL-1.5-8B2026.0633
VideoChat-Flash-7B2026.0632.5
Eagle-X5-7B2026.0631
GLM-4V-9B2026.0630
MiniCPM-V-2.6-8B2026.0629.8
CoPE-VideoLM-7B2026.0628.3
LLaVA-Video-7B2026.0624.9
InternVL2-8B2026.0624.6