WIP: Add vLLM support to ChatQnA + DocSum Helm charts #610
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
|
Setting as draft. I have tested that DocSum with Gaudi vLLM works, and that ChatQnA Helm chart can be installed, but due to v1.1 image pulls currently taking so long in my test node, I haven't been able test ChatQnA with Gaudi vLLM properly yet. vLLM CPU version testing would also be needed before merging this (I'm hoping somebody else here could check at least DocSum with CPU vLLM). |
|
CI issues:
|
|
This overlaps partly with #403. |
3 tasks
Signed-off-by: Eero Tamminen <[email protected]>
Otherwise service throws an exception due to None variable value. Signed-off-by: Eero Tamminen <[email protected]>
|
Added HPA support for ChatQnA / vLLM. |
Signed-off-by: Eero Tamminen <[email protected]>
Signed-off-by: Eero Tamminen <[email protected]>
For now vLLM replaces just TGI, but as it supports also embedding, also TEI-embed/-rerank may be replaceable later on. Signed-off-by: Eero Tamminen <[email protected]>
Signed-off-by: Eero Tamminen <[email protected]>
Signed-off-by: Eero Tamminen <[email protected]>
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Outdated suggestions cannot be applied.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.
Description
Add vLLM support to ChatQnA + DocSum Helm app charts.
Similarly to how it's already done in Agent component, these have now
tgi.enabledandvllm.enabledflags for selecting which LLM will be used.Notes:
llm-docsum-vllmwrapper image from GenAIComps repo, as that image is currently missing from DockerHubIssues
Fixes #608 partially.
Type of change
New dependencies
opea/llm-docsum-vllm:latestimage is currently missing from CI & DockerHub registries: https://github.com/opea-project/GenAIComps/tree/main/comps/llms/summarization/vllm/langchain/Tests
Manual testing on top of "main" HEAD / v1.1 images.