List your company

Video Question Answering on EgoPlan-Bench

43.06AccuracyGPT-5.2
23.882428.861233.8438.8188Jun 15, 2026

Evaluation results

MethodLinks
GPT-5.2Encoding=Uniform-8, Variant=Plain2026.0643.06
Gemini 3 FlashEncoding=Uniform-8, Variant=Plain2026.0637.96
Gemini 3 FlashEncoding=Cascade, Variant=Seed2026.0630.8
FullEvo (Cat.)Backbone=Gemini 3 Flash, Encoding=Cascade2026.0630.04
Gemini 3 FlashEncoding=Cascade, Variant=+Evolve2026.0629.93
FullEvo (Guide)Backbone=GPT-5.2, Encoding=Cascade2026.0629.39
FullEvo (Guide)Backbone=Gemini 3 Flash, Encoding=Cascade2026.0628.85
FullEvo (Cat.)Backbone=GPT-5.2, Encoding=Cascade2026.0628.85
GPT-5.2Encoding=Cascade, Variant=Seed2026.0628.74
GPT-5.2Encoding=Cascade, Variant=Plain2026.0628.42
Gemini 3 FlashEncoding=Cascade, Variant=+SkillMemCat2026.0628.31
GPT-5.2Encoding=Cascade, Variant=+Evolve2026.0628.31
GPT-5.2Encoding=Cascade, Variant=+SkillMemCat2026.0628.09
Gemini 3 FlashEncoding=Cascade, Variant=Plain2026.0624.62