Adding support for VLLM server #495
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
bolasim
reviewed
Aug 1, 2023
bolasim
approved these changes
Aug 1, 2023
bolasim
approved these changes
Aug 3, 2023
| @@ -71,6 +71,41 @@ def create_tgi_build_dir(config: TrussConfig, build_dir: Path): | |||
| supervisord_filepath.write_text(supervisord_contents) | |||
|
|
|||
|
|
|||
| def create_vllm_build_dir(config: TrussConfig, build_dir: Path): | |||
| server_endpoint_config = { | |||
There was a problem hiding this comment.
nit: constant should be all caps
| ) | ||
| nginx_template = read_template_from_fs(TEMPLATES_DIR, "vllm/proxy.conf.jinja") | ||
|
|
||
| dockerfile_content = dockerfile_template.render(hf_access_token=hf_access_token) |
There was a problem hiding this comment.
we should write a helpful function at this point. we've copied like code like 7 timesnow. I can do that in a followup.
aspctu
added a commit
that referenced
this pull request
Aug 3, 2023
* Cleanup old asyncio threadpool settings. * Add `version.txt` into build to prevent cache migrations (#490) * added hf token to dockerfile * add prints * move hf token to the part where its needed * add version.txt * added version.txt * remove unnecessary files * adding files back * reverting * Adding `allow_patterns` and `ignore_patterns` to the Hugging Face cache. (#480) * remove assert * both ignore and allow working * fix for verbose * remove prints * Fix streaming issues. * Added documentation for HF caching (#492) * added hf_cache docs * fix ticks * Update configuration.md * clean up Truss docs (#491) * clean up Truss docs * fix links and lints * lint * Comment updates. * Bump version. * Controlling supervisord retries (#496) * update live reload docs (#494) * update live reload docs * remove leading space * remove dead links from flan-t5 readme * Update README.md * Removing extraneous file (#498) * Enable Hugging Face secrets during build from Truss (#499) * added hf token to dockerfile * add prints * move hf token to the part where its needed * successfully mounting secrets * update cache warmer to grab secret * match data dir copy * bump pyproject * add os to cache_warmer * bump pyproject * add is_trusted * revert version * change names to be hf_access_token * rename is_trusted and use Path * Adding support for VLLM server (#495) --------- Co-authored-by: Sidharth Shanker <sid.shanker@baseten.co> Co-authored-by: Varun Shenoy <vnshenoy@stanford.edu> Co-authored-by: Philip Kiely - Baseten <98474633+philipkiely-baseten@users.noreply.github.com> Co-authored-by: joostinyi <63941848+joostinyi@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds support to deploy models with the vLLM server.