fix(ollama): normalize stray system messages to satisfy strict templates - #85
Merged
Conversation
Qwen-style Ollama chat templates enforce a single system message pinned to position 0 and raise 'System message must be at the beginning.', surfaced as an opaque 400 'Unable to generate parser for this template'. Lettuce injects depth-positioned system entries mid-conversation, which tripped this guard. Fold the leading system run into one opening block and demote any later system message to user, preserving content and position.
Coalesce demoted mid-conversation system entries with adjacent user turns instead of emitting consecutive user messages. This keeps strict Ollama templates valid while retaining injected prompt content in conversation order.
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Qwen-style Ollama chat templates enforce a single system message pinned to position 0 and raise 'System message must be at the beginning.', surfaced as an opaque 400 'Unable to generate parser for this template'. Lettuce injects depth-positioned system entries mid-conversation, which tripped this guard.
Fold the leading system run into one opening block and demote any later system message to user, preserving content and position.