Repository navigation
fix(api): clarify when to omit completion choice count - #4298
Conversation
Codex Review Summary: ✅ CompletedThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
markstuart-oai
left a comment
There was a problem hiding this comment.
Please scope the newer-model guidance to Chat Completions.
Castiron-Internal-PR: openai/openai-python-internal#444 Castiron-Source-SHA: f2b620726de993cf62296f123a823cc3c2f80382 Castiron-Public-Base-SHA: bb2d8c5
18ed7b6 to
df67e19
Compare
jbeckwith-oai
left a comment
There was a problem hiding this comment.
Approved after static review of df67e1969a95261c5c0403936320069a0b2b3192 with an independent reviewer. Verified the complete public diff and changed-file contents against the approved internal revision, including schema descriptions, generated documentation, and manifest hashes. The current wording addresses the existing review request: the newer-model explicit-n warning is limited to Chat Completions; legacy Completions retain the one-completion default guidance and token-quota warning. Parameter types, defaults, serialization, and runtime behavior are unchanged. No additional actionable findings.
Hosted checks at publication: 15 success, 9 skipped, 2 in_progress. Unfinished CI remains tracked. No candidate code or local tests executed.
markstuart-oai
left a comment
There was a problem hiding this comment.
I’ve been checking the regenerated completion-count docs. The legacy endpoint now documents one completion per prompt when n is omitted, and the newer-model guidance is confined to Chat Completions.
codex://threads/01a1c7cc-3ae0-7900-8710-5a1fbeb062d5
Summary
Clarifies the choice-count guidance for Chat Completions and Completions.
Changes