Skip to content

[Docs] Cover GPT-6 family reasoning efforts on the OpenAI provider page - #63

Draft
zoomote[bot] wants to merge 1 commit into
mainfrom
feature/openai-gpt-6-docs-2n7yvyumlorye
Draft

zoomote[bot] wants to merge 1 commit into
mainfrom
feature/openai-gpt-6-docs-2n7yvyumlorye

Conversation

@zoomote

@zoomote zoomote Bot commented Sep 30, 2026

Copy link
Copy Markdown

​Opened on behalf of @taltas. Follow up by mentioning @roomote, in the web UI, or in Discord.

What changed

docs/providers/openai.md now covers the GPT-6 family alongside GPT-5: the intro mentions both families, and the Reasoning Effort Control section documents the GPT-6 effort levels (low, medium default, high, xhigh, max) and notes that most GPT-6 models (GPT-6 Sol, GPT-6 Luna) also support none, while GPT-6.1 Sol and GPT-6 Astra require an explicit effort and do not support none/minimal.

Why this change was made

Paired with the Zoo-Code PR adding gpt-6.1-sol to the OpenAI model catalogs, so the provider documentation reflects the GPT-6 family users can now select and how reasoning effort behaves for the new models.

Impact

Docs-only change; no behavior impact. Readers of the OpenAI provider page get accurate GPT-6 reasoning-effort guidance. No docs pages list per-model pricing (the page defers to OpenAI's pricing page), so no pricing tables needed updating, and no docs UI screenshots show model pickers affected by this change.

Related PRs

Screenshots

OpenAI provider docs page showing the new GPT-6 Models reasoning-effort section

Pair with the Zoo-Code change adding gpt-6.1-sol to the OpenAI model
catalogs: mention the GPT-6 family alongside GPT-5 and document the
GPT-6 reasoning effort levels (low through max), including that
GPT-6.1 Sol and GPT-6 Astra require an explicit effort and do not
support none/minimal.

This branch has not been deployed

No deployments
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