fix: enable native streaming for Azure o4 models

[colesmcintosh]

Title

Enable native streaming for Azure o4 models

Relevant issues

Fixes #12150

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/litellm/ directory, Adding at least 1 test is a hard requirement - see details
• I have added a screenshot of my new test passing locally
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem

Type

🐛 Bug Fix

Changes

• Updated should_fake_stream method in litellm/llms/azure/chat/o_series_transformation.py to return False for o4 models (including o4-mini)
• Azure o4 models now support native streaming per Microsoft documentation
• Added tests to verify o4 streaming behavior in tests/llm_translation/test_azure_o_series.py

Test Results

Tests pass successfully:

tests/llm_translation/test_azure_o_series.py::test_should_fake_stream_for_o4_models PASSED
tests/llm_translation/test_azure_o_series.py::test_azure_o3_streaming PASSED
tests/llm_translation/test_azure_o_series.py::test_azure_o4_streaming PASSED

Changes Summary

1. Modified should_fake_stream to check for both "o3" and "o4" in model names
2. Updated docstring to reflect that both O3 and O4 support native streaming
3. Added comprehensive tests for o4 streaming behavior

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
litellm	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Jun 30, 2025 3:33pm