List your company

Mathematical Reasoning on GSM8K PRO

95AccuracyDIN-ICL
79.81683.75887.791.642Apr 7, 2026

Evaluation results

MethodLinks
DIN-ICLModel Series=Qwen-3, Parameters=32B2026.0495
SET-BSRModel Series=Qwen-3, Parameters=32B2026.0494.6
DIN-ICLModel Series=Qwen-2.5, Parameters=14B2026.0494
SET-BSRModel Series=Qwen-2.5, Parameters=14B2026.0493.2
SET-BSRModel Series=Gemma-3, Parameters=27B2026.0493.1
DIN-ICLModel Series=Qwen-3, Parameters=8B2026.0493
DIN-ICLModel Series=Gemma-3, Parameters=27B2026.0492.8
DIN-ICLModel Series=Gemma-3, Parameters=12B2026.0492.7
SET-BSRModel Series=Qwen-3, Parameters=8B2026.0492.2
SET-BSRModel Series=Gemma-3, Parameters=12B2026.0492.1
DIN-ICLModel Series=Qwen-2.5, Parameters=32B2026.0492.1
SET-BSRModel Series=Qwen-2.5, Parameters=32B2026.0491.6
SET-BSRModel Series=Qwen-3, Parameters=14B2026.0491.1
DIN-ICLModel Series=Qwen-3, Parameters=14B2026.0490.9
DIN-ICLModel Series=Qwen-2.5, Parameters=7B2026.0490.3
SET-BSRModel Series=Qwen-2.5, Parameters=7B2026.0489.2
DIN-ICLModel Series=LLaMA-3.1, Parameters=8B2026.0481.6
SET-BSRModel Series=LLaMA-3.1, Parameters=8B2026.0480.4