VerlTool/torl_qwen2.5-math-7b-grpo-n16-b128-t1.0-lr1e-6acc-only-global_step_200
TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Aug 17, 2025Architecture:Transformer Cold
Loading preview...
Loading preview...