fix(llm): support 'reasoning' field from Ollama in streaming responses#843
Open
abdelhadi703 wants to merge 1 commit intoQwenLM:mainfrom
Open
fix(llm): support 'reasoning' field from Ollama in streaming responses#843abdelhadi703 wants to merge 1 commit intoQwenLM:mainfrom
abdelhadi703 wants to merge 1 commit intoQwenLM:mainfrom
Conversation
Ollama sends 'delta.reasoning' for Qwen3 models with thinking content, but the streaming code only checked for 'delta.reasoning_content' (used by vLLM/Qwen). This caused thinking content to be silently dropped during streaming. Use getattr() to check both field names, supporting both 'reasoning_content' (vLLM/Qwen) and 'reasoning' (Ollama) in both delta_stream and full accumulation modes. Fixes QwenLM#789
abdelhadi703
force-pushed
the
fix/ollama-streaming-reasoning
branch
from
077986e to
d8218b7
Compare
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Fix streaming responses from Ollama where thinking content is silently lost.
Problem: Ollama sends
delta.reasoningfor Qwen3 models with thinking content, but the streaming code only checked fordelta.reasoning_content(used by vLLM/Qwen API).Fix: Use
getattr()to check both field names —reasoning_content(vLLM/Qwen) andreasoning(Ollama) — in both streaming modes (delta yield and full accumulation).Changes
qwen_agent/llm/oai.py: Updated_chat_stream()to handle both field namesTest
Verified that both
reasoning_content(vLLM) andreasoning(Ollama) are now correctly captured.Contribution by abdelhadisalmaoui0909@outlook.fr