fix: enable mergeToolResultText for LM Studio and OpenAI-compatible providers #11076
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
…roviders Enable the mergeToolResultText option for LM Studio and BaseOpenAiCompatibleProvider to align their message formatting behavior with the Z.ai provider. This change helps prevent models like GLM4.5 from getting stuck in repeated file read loops by merging text content after tool_results into the last tool message instead of creating separate user messages that can disrupt the model conversation flow. Fixes #11071
Contributor
Author
Review complete. No issues found. The changes correctly enable Mention @roomote in a comment to request specific changes to this pull request or fix all unresolved issues. |
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Outdated suggestions cannot be applied.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
This PR attempts to address Issue #11071 by enabling the
mergeToolResultTextoption for LM Studio and BaseOpenAiCompatibleProvider to align their message formatting behavior with the Z.ai provider.Problem
GLM4.5 and other models get stuck in repeated file read loops when using LM Studio or OpenAI-compatible endpoints. This is because text content after tool_results (like
environment_details) creates separate user messages that can disrupt the model's conversation flow.Solution
Enable the
mergeToolResultText: trueoption when callingconvertToOpenAiMessages()in:src/api/providers/lm-studio.tssrc/api/providers/base-openai-compatible-provider.tsThis matches the behavior of the Z.ai provider, which already uses this option to preserve conversation flow by merging text content after tool_results into the last tool message.
Changes
mergeToolResultText: truetoconvertToOpenAiMessages()callmergeToolResultText: truetoconvertToOpenAiMessages()callTesting
Related
Feedback and guidance are welcome.
Important
Enable
mergeToolResultTextfor LM Studio and OpenAI-compatible providers to prevent conversation flow disruption.mergeToolResultText: trueinconvertToOpenAiMessages()forLM StudioandBaseOpenAiCompatibleProviderto prevent conversation flow disruption.lm-studio.ts: UpdatedcreateMessage()to usemergeToolResultText: true.base-openai-compatible-provider.ts: UpdatedcreateStream()to usemergeToolResultText: true.This description was created by
for 74deee8. You can customize this summary. It will automatically update as commits are pushed.