Our new X account is live! Follow @wizwand_team for updates

Video Understanding on TempCompass MCQ (test)

82.8Accuracy

Gemini 3 Pro

Updated 3d ago

Evaluation Results

Method	Links
Gemini 3 Pro 2026.01		82.8
Gemini 3 Pro 2026.02		82.8
Gemini 2.5 Pro 2026.01		81.9
Gemini 2.5 Pro 2026.02		81.9
GPT-5 2026.01		80.4
GPT-5 2026.02		80.4
Gemini 2.5 Flash 2026.01		80.2
Keye-VL-1.5-8B 2026.01		75.5
GPT-5 mini 2026.01		74.9
Eagle2.5-8B 2026.01		74.4
Qwen3-VL-8B 2026.01		74.3
Molmo2-8B 2026.01		73.4
Molmo2-O-7B 2026.01		73
Claude Sonnet 4.5 2026.01		72.8
Molmo2-4B 2026.01		72.8
Claude Sonnet 4.5 2026.02		72.8
MiniCPM-V-4.5-8B 2026.01		72.7
PLM-8B 2026.01		72.7
GLM-4.1V-9B 2026.01		72.3
Qwen3-VL-4B 2026.01		70.8
InternVL3.5-8B 2026.01		70.3
VideoChat-Flash-7B 2026.01		69.4
PLM-3B 2026.01		69.3
InternVL3.5-4B 2026.01		68.8
CoPE-VideoLM 2026.02		68.4
IXC-2.5-7B 2026.02		67.1
LLaVA-Video-7B 2026.01		66.6
LLaVA-Video-7B 2026.02		66.6
InternVL2-8B 2026.02		65.3
Apollo-7B 2026.02		64.9
LLaVA-OV-7B 2026.02		64.8
LongVA-7B 2026.02		56.9
VideoChat2-7B 2026.02		45.5