Does the R1.1 model internally implement a VLLM-based inference framework?

#7
by 026jzz - opened

I'm working on the vllm-omni adaptation for this model, and a friend reminded me that you guys seem to have already implemented it, so I'm asking here.
btw, your presentation is amazing, I watched the Live recording through bilibili

yes, please see https://github.com/stepfun-ai/Step-Audio-R1 for further information.

Sign up or log in to comment