AI BriefingOpenAI Codex Adds Support for 'Persistent Reasoning Effort' Feature
Key point
OpenAI Codex adds a 'persistent' option to the reasoning-effort protocol and supports it in the TUI and SDK.
Details
OpenAI Codex has added the persistent option to the reasoning-effort protocol and TypeScript SDK types. This change displays the persistent reasoning capabilities advertised by models as 'Persistent' in the Terminal User Interface (TUI), while maintaining the 'persistent' value in local settings but sending the 'disabled' value to the Responses API.
Key Changes
- Protocol and SDK Updates: Added the
persistentfield to the reasoning-effort protocol and TypeScript SDK types. - TUI Display Changes: Clearly displays the persistent reasoning capabilities supported by models as 'Persistent' in the TUI.
- Settings and API Integration: Preserves the
persistentvalue in local configuration files but converts it to send thedisabledwire value to the Responses API during actual API requests.
Test Coverage
The feature was tested across various scenarios, including parsing and serialization, request transformation, remote model requests, TUI reasoning selector, and TypeScript CLI argument passing.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
OpenAI Prepares New $500/Month ChatGPT Pro Max Plan with 'Fastest' Inference
Atlas: Open-Source Source Control Tool for Multi-Agent Coding Workflows
OpenAI Discloses AI Agents Accessed US Government Websites During TrainingOpenAI Pauses Training of Most Capable Models After Agent Exploits DNS Gap to Reach External Chatbot