fix: support azure o3 model family for fake streaming workaround #8162
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
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
litellm/utils.py
Outdated
| @@ -5997,7 +5997,7 @@ def get_provider_chat_config( # noqa: PLR0915 | |||
| ): | |||
| return litellm.AI21ChatConfig() | |||
| elif litellm.LlmProviders.AZURE == provider: | |||
| if litellm.AzureOpenAIO1Config().is_o1_model(model=model): | |||
| if litellm.AzureOpenAIO1Config().is_reasoning_model(model=model): | |||
There was a problem hiding this comment.
given the refactor will probably be to the o-series can we rename the helper util something like - is_o_series_model ?
it would just be clearer what it's really trying to do
testing here @byrongrogan the basellmchattest goes through a series of common scenarios for the model it's fine - i can take care of this part, since this will probably require us to have an o3 deployment on azure |
b553028
into
BerriAI:litellm_dev_02_01_2025_p1
2 checks passed
krrishdholakia
added a commit
that referenced
this pull request
Feb 1, 2025
krrishdholakia
added a commit
that referenced
this pull request
Feb 1, 2025
* fix: support azure o3 model family for fake streaming workaround (#8162) * fix: support azure o3 model family for fake streaming workaround * refactor: rename helper to is_o_series_model for clarity * update function calling parameters for o3 models (#8178) * refactor(o1_transformation.py): refactor o1 config to be o series config, expand o series model check to o3 ensures max_tokens is correctly translated for o3 * feat(openai/): refactor o1 files to be 'o_series' files expands naming to cover o3 * fix(azure/chat/o1_handler.py): azure openai is an instance of openai - was causing resets * test(test_azure_o_series.py): assert stream faked for azure o3 mini Resolves #8162 * fix(o1_transformation.py): fix o1 transformation logic to handle explicit o1_series routing * docs(azure.md): update doc with `o_series/` model name --------- Co-authored-by: byrongrogan <47910641+byrongrogan@users.noreply.github.com> Co-authored-by: Low Jian Sheng <15527690+lowjiansheng@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.
Support o3 model family for fake streaming workaround from Azure
o3-miniis already available on Azure, but sadly it looks like it doesn't support streaming, just like theo1models.So this PR updates the logic for the fake streaming workaround to also support the
o3family. Note that there's a wider refactor to generalize the name ofAzureOpenAIO1Configthat I'm not making here.Relevant issues
Fixes #8158
Type
🐛 Bug Fix
Changes
o3*to the list of models for the Azure fake streaming workaround[REQUIRED] Testing - Attach a screenshot of any new tests passing locally
I didn't see any obvious tests in this area, so manually tested the fake streaming is now working with
o3-mini,o1-mini, as well as non-streaming ono3-mini. Also validated that real streaming is still working withgpt-4o.