Fix Ollama GPT-OSS streaming with 'thinking' field #13375
+85
−1
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
- Handle chunks containing 'thinking' field with empty 'response' - Treat these as intermediate chunks that don't contain user content - Add comprehensive tests for chunk parsing scenarios - Resolves APIConnectionError for GPT-OSS model streaming Fixes BerriAI#13340
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
Fixed MyPy type error in Ollama completion transformation where finish_reason was set to None instead of expected string type. Changed finish_reason=None to finish_reason="" to match GenericStreamingChunk TypedDict requirements. Also updated corresponding test to expect empty string instead of None.
| @@ -459,6 +459,15 @@ def chunk_parser(self, chunk: dict) -> GenericStreamingChunk: | |||
| finish_reason="stop", | |||
| usage=None, | |||
| ) | |||
| elif "thinking" in chunk and not chunk["response"]: | |||
There was a problem hiding this comment.
@colesmcintosh can we return a ModelResponseStream instead?
This way we can include the reasoning content in the response, and allow it to be displayed on chat ui's like OpenWebUI - See openrouter implementation -
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Outdated suggestions cannot be applied.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.
Title
Fix Ollama GPT-OSS streaming with 'thinking' field
Relevant issues
Fixes #13340
Pre-Submission checklist
Please complete all items before asking a LiteLLM maintainer to review your PR
tests/litellm/directory, Adding at least 1 test is a hard requirement - see detailsmake test-unitType
🐛 Bug Fix
Changes
Problem
Ollama GPT-OSS models were failing with
APIConnectionError: Unable to parse ollama chunkwhen streaming responses contained a'thinking'field with empty'response'. The chunk parser didn't handle this specific case, causing streaming to fail.Solution
'thinking'field with empty'response'Code Changes
OllamaTextCompletionResponseIterator.chunk_parser()inlitellm/llms/ollama/completion/transformation.py"thinking" in chunk and not chunk["response"]GenericStreamingChunkfor these intermediate chunksTests Added
test_chunk_parser_with_thinking_field(): Tests the exact problematic chunk from the issuetest_chunk_parser_normal_response(): Ensures normal chunks still worktest_chunk_parser_done_chunk(): Verifies done chunks work correctlyVerification
Tested with the exact chunk from the error:
{'model': 'gpt-oss:20b', 'created_at': '2025-08-06T14:34:31.5276077Z', 'response': '', 'thinking': 'User', 'done': False}Result: Successfully parsed without errors, returns empty text chunk allowing stream to continue.