fix: stop reasoning-only loops for GLM/minimax #111
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Outdated suggestions cannot be applied.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.
Problem
Some models (reported with
z-ai/glm-4.7:freeandminimax/minimax-m2.1:free) can stream only reasoning (<think>/reasoning deltas) and finish the step withfinish: \"unknown\".SessionPrompt.loopkeeps looping onunknown, causing an infinite repetition.Original issue: #104
Fix
MessageReasoningStuckError.SessionProcessor, track whether a step produced any non-reasoning output (text/tool parts).unknown), mark the message withMessageReasoningStuckErrorand stop the loop.Tests
bun run typecheckbun test(packages/opencode)Closes #104