shengjia-toronto/sac-gspo-cl5e3-drgrpo-llama32-3b-deepscaler-step881-best-pass1-16.34-8xH200

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:3.2BQuant:BF16Ctx Length:32kPublished:May 22, 2026Architecture:Transformer Warm

Loading preview...