Rework Backend to Native HTTP Requests and Enhance API Compatibility & Performance #91
Conversation
This file contains hidden or 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
…hey become serializable
…hey become serializable
…http/2 support. Additionally, add in support for multi modal requests (needs further enablement in the rest of the system for future TODO). Still needs testing and test fixes
markurtz
force-pushed
the
openai_backend_fixes
branch
from
24f753a to
9e4bc84
Compare
This was referenced Mar 10, 2025
Closed
sjmonson
approved these changes
Mar 11, 2025
There was a problem hiding this comment.
Sorry this took so long to review. My testing showed that there was a big regression in the time between each request for all scenarios. But I believe its just due to the start and stop timing being moved closer to the actual start and stop of the request.
Otherwise LGTM
Thanks @sjmonson! I think that's likely. Also, provided we're not seeing a severe regression, I would say let's focus on optimizing the perf with #96 |
markurtz
changed the title
markurtz
pushed a commit
that referenced
this pull request
May 7, 2025
Prior to the `openai_server` -> `openai_http` refactor (#91), we were using the `extra_query` parameter [in the OpenAI client](https://github.com/openai/openai-python/blob/fad098ffad7982a5150306a3d17f51ffef574f2e/src/openai/resources/models.py#L50) to send custom query parameters to the OpenAI server in requests made by guidellm. This PR adds that parameter to the new `OpenAIHTTPBackend`, making it possible to add custom query parameters that are included in every request sent to the server.
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Outdated suggestions cannot be applied.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.
Summary
This PR restructures the backend of
guidellmto utilize native HTTP requests withhttpxand HTTP/2 standards, significantly improving performance and scalability. It aligns the backend interface closely with the OpenAI API specifications, enabling smoother integration paths for future expansions, including multi-backend support.Details
httpx, enabling HTTP/2 support for optimized performance.base.py,load_generator.py.backend.py,scheduler/backend_worker.py, and structured response handling inresponse.py.config.py) to better support environment-based settings.Testing
Related Issues