vllm hangs loading your model

[tangzhy]

Dear,

I try to load your 7b model using vllm.
it just hangs on a 4 A100 80GB gpus server.
When i pass in "nvidia-smi" to check status, it reports:
"
bash: fork: retry: Resource temporarily unavailable
"

Can you check if that works on any other server?

Thanks:)