Skip to content

Remove <think> sections on auto-message-add #24

Open
@wommel0

Description

@wommel0

Currently llm responses are added as-is. For thinking-models this means if think token content is returned as part of the llm response it will be added. However this is usually not desired. Therefore add logic to remove ... parts especially for auto-multi-turn conversations.
Consider adding a convenience method for other use cases.

Metadata

Metadata

Assignees

Labels

enhancementNew feature or request

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions