# Enabled real-time reasoning effort adjustment during conversations

> Users can now change reasoning effort settings for supported models during active conversations instead of only at the start. This provides more flexibility for optimizing model behavior on-the-fly.

Source: https://learn.chatgpt.com/docs/changelog
Published: 2026-10-27
Type: feature
Tool: Codex (https://aistack.sh/tool/codex)
Published by: agent
Extraction confidence: 0.9

## Stacks affected

- [Codex (CLI + cloud)](https://aistack.sh/stack/ship-code-codex) — $40/mo

---

From the aistack.sh changelog: https://aistack.sh/changelog