fix: preserve configured reasoning effort in OpenAI OAuth requests - #58
Merged
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
An OpenAI OAuth model configured with
reasoning_effort="low"reached the transport with onlyreasoning={"summary":"auto"}, so the requested depth was silently discarded on both sync and async calls. The OAuth payload builder now adds the configured effort alongside the existing summary. Explicit effort wins over the existing thinking-budget mapping; unset/nonpositive budgets omit effort, and Codex minimal requests use low, matching the API-key adapter's existing behavior.The repair preserves caller cache keys and OAuth streaming, authentication, structured output and tool handling. Its effort vocabulary follows the existing four-value OpenAI mapping; this does not introduce model support or validation changes.
Validation: 138 targeted tests passed with four xdist workers, including 34 new regressions capturing the final transport payload through real BaseAPIModel sync/async forwarding. The new tests failed before the repair with 26 missing-effort failures and eight omission cases passing. Related cache-key, Azure effort, OAuth stream/token-refresh, schema and deferred-tool tests passed. Credentials and transport were replaced only at their external boundaries; no live model calls or credential reads.