VLLM takes alot of time to start, can we use the docker image? #3872
Unanswered
Stealthwriter
asked this question in
Q&A
Replies: 2 comments
-
you should be good with picking a vllm container that suites your needs and load it as described in https://skypilot.readthedocs.io/en/latest/examples/docker-containers.html#docker-containers-as-runtime-environments |
Beta Was this translation helpful? Give feedback.
0 replies
-
We have an example using vllm docker image here: https://github.com/skypilot-org/skypilot/blob/master/llm/vllm/serve-openai-api-docker.yaml Note that runpod has some limitations and this yaml may not work. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi,
I'm trying to use sky serve with the docker image of vllm, and my cloud provider is runpod, which YAML should I use?
Beta Was this translation helpful? Give feedback.
All reactions