parkjo/Qwen2.5-Math-7B_grpo_adv_rollout_8_USE_KL_True_20260513_123239_step580
TEXT GENERATIONConcurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:May 15, 2026Architecture:Transformer Featherless Exclusive Cold
Loading preview...