On: vLLM v0.28.0

[SillyUsername]

I just wish they'd support Pascal :(

Nvidia might have given up support but it doesn't mean vllm have to (llama.cpp didn't).

[ta93754829 → SillyUsername]

Pascal? The programming language?

[lightdot → ta93754829]

In this case, Pascal as a Nvidia's codename for a GPU microarchitecture.

https://en.wikipedia.org/wiki/Pascal_(microarchitecture)

I guess many would like to use stock vLLM to run inference on affordable P100...

[SillyUsername → ta93754829]

Sorry I should have qualified it as Nvidia Pascal architecture.