Upgrade OpenAI SDK #2384
Merged
Upgrade OpenAI SDK #2384
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
yifanmai
force-pushed
the
yifanmai/fix-upgrade-openai-sdk
branch
from
57b7aca
to
b28d827
Compare
JosselinSomervilleRoberts
approved these changes
Feb 20, 2024
| @@ -43,25 +41,17 @@ def __init__( | |||
| cache_config: CacheConfig, | |||
| api_key: Optional[str] = None, | |||
| org_id: Optional[str] = None, | |||
| base_url: Optional[str] = None, | |||
There was a problem hiding this comment.
Is this base_url set anywhere or should this be specified somewhere? Since in old requests, the URL was wet by default to https://api.openai.com/v1, I am wondering if old requests that are not explicitly this will now fail? Or is the behaviour of OpenAI to set it as this if it is None?
There was a problem hiding this comment.
OpenAI sets the default here. It's probably a good idea to let OpenAI do it, and not duplicate this in our own code.
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Outdated suggestions cannot be applied.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.
Migrate to OpenAI SDK v1.0.0, using the guide and CLI tool here.
I manually tested this by sending requests and checking that the
RequestResultlooked acceptable for the following models:gpt-3.5-turbo-instruct(completion),gpt-3.5-turbo-0613(chat),text-embedding-ada-002(embedding), anddall-e-2(image).The only behavior change is that we now use "model" instead of "engine" in the request to completion and embedding models, because "engine" is deprecated. This means that new requests will not match existing cache keys for completion and embedding models. Chat models were already using "model" and are unaffected.
Also fixes an existing issue where
ModerationAPIClientis instantiated eagerly instead of lazily inServerService.Also add a configurable
base_url, which should allow the client to be used with vLLM.Addresses #1997.