Possible to run on six RTX Pro 6000 Blackwell with vLLM oder SGLang?

#2
by FabianHeller - opened

Is it possible to run this model with 6 RTX Pro 6000 Blackwell GPUs? I think generally it should work with a combination of --tensor-parallel-size 2 and --pipeline-parallel-size 3, but I am not sure.

would be interesting to see if it does

Not on the latest vllm:
NotImplementedError: Pipeline parallelism is not supported for this model. Supported models implement the SupportsPP interface.

Sign up or log in to comment