OpenAI Codex Update Adds Persistent Reasoning Support But Raises Compatibility Concerns for Custom Providers
Summary
OpenAI Codex receives a major update adding persistent reasoning support across its protocol and SDK, but a newly flagged compatibility issue warns that custom provider configurations using a 'persistent' effort value may be silently rewritten to 'disabled', potentially breaking non-OpenAI integrations.
Key Points
- A new pull request merged into the OpenAI Codex main branch on August 26, 2026 adds support for persistent reasoning effort across the protocol, TypeScript SDK types, and local configuration.
- The update displays model-advertised persistent reasoning as 'Persistent' in the TUI while internally translating the value to the Responses API's 'disabled' wire format to maintain compatibility.
- An automated code review flags a potential issue where custom provider configurations using a literal 'persistent' effort value may now be unexpectedly rewritten to 'disabled', potentially altering behavior for non-OpenAI providers.