Skip to content

fix(llm, openai): Separate reasoning items - #1165

Merged
JeanMertz merged 1 commit into
mainfrom
openai-reasoning
Sep 16, 2026
Merged

JeanMertz merged 1 commit into
mainfrom
openai-reasoning

Conversation

@JeanMertz

Copy link
Copy Markdown
Collaborator

Consecutive OpenAI reasoning items render as separate paragraphs during streaming and when replaying newly recorded conversations. Summary indices restarting at zero no longer glue adjacent headings together.

Insert missing separation before the next reasoning text, preserving existing line breaks and leaving final items untouched. This avoids synthetic trailing text consuming the reasoning display's truncation budget. Keep segmentation provider-local so Anthropic reasoning split mid-word still renders continuously.

Consecutive OpenAI reasoning items render as separate paragraphs during
streaming and when replaying newly recorded conversations. Summary
indices restarting at zero no longer glue adjacent headings together.

Insert missing separation before the next reasoning text, preserving
existing line breaks and leaving final items untouched. This avoids
synthetic trailing text consuming the reasoning display's truncation
budget. Keep segmentation provider-local so Anthropic reasoning split
mid-word still renders continuously.

Signed-off-by: Jean Mertz <git@jeanmertz.com>
@JeanMertz
JeanMertz merged commit d2a7ee4 into main Sep 16, 2026
21 checks passed
@JeanMertz
JeanMertz deleted the openai-reasoning branch September 16, 2026 17:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant