List your company

Multimodal Vision Evaluation on BabyVision

43.23AccuracyRSA
41.493241.944142.39542.8459Apr 9, 2026

Evaluation results

MethodLinks
RSAStrategy=RSA, Model 1=—, Model 2=Kimi-2.5-Thinking2026.0443.232.051
SQUEEZE EVOLVE (p=0)Strategy=SQUEEZE EVOLVE (p=0), Model 1=Kimi-2.5-Instant, Model 2=Kimi-2.5-Thinking2026.0441.560.812.5