-
Notifications
You must be signed in to change notification settings - Fork 36
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Adding vllm speculative decoding example #317
base: main
Are you sure you want to change the base?
Conversation
base_image: | ||
image: nvcr.io/nvidia/pytorch:23.11-py3 | ||
python_executable_path: /usr/bin/python3 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
just cuious why we need this base image? Can you add a cooment?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Without this base image the build does not succeed. The baseten base image does not have nvcc
, which is required for the developer build of vLLM.
tensor_parallel: 1 | ||
max_num_seqs: 16 | ||
model_name: vLLM Speculative Decoding | ||
python_version: py310 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
drop python version if using base image
python_executable_path: /usr/bin/python3 | ||
build_commands: [] | ||
environment_variables: | ||
HF_TOKEN: "" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
why here over secrets?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
vLLM reads only this specific environment variable for the access token. It doesn't work with secrets
No description provided.