shengjia-toronto/DeepScaleR-1.5B-16k-GAPO-GSPO-NoKL-Step175-AIME24-40pct

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:1.5BQuant:BF16Ctx Length:32kPublished:May 23, 2026Architecture:Transformer Warm

Loading preview...