fix: prevent content truncation during OpenRouter reasoning-to-text transition #891
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
…ransition When streaming models that emit thinking tokens (e.g. qwen/qwen3-8b), OpenRouter sends transition chunks where the `reasoning` field is an empty string and `content` contains the start of the actual response. Because `hasReasoningDelta()` checks for field existence via `isset()`, these chunks match even when reasoning is empty. The unconditional `continue` then skips the content extraction, truncating the first characters of the response. Move `continue` inside the `$reasoningDelta !== ''` check so that chunks with empty reasoning fall through to content processing.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When streaming models that emit thinking tokens (e.g.
qwen/qwen3-8bvia OpenRouter), transition chunks contain an emptyreasoningfield alongside the start of the actualcontent.hasReasoningDelta()usesisset(), which returnstruefor empty strings. The unconditionalcontinueon line 187 then skips content extraction for these chunks, truncating the first characters of the response.Fix: Move
continueinside the$reasoningDelta !== ''check so that chunks with empty reasoning fall through to content processing.