How to change reasoning effort in Codex CLI: model_reasoning_effort values and one-off overrides
Originally published at https://aicoding-guide.com . Codex feels slow, or it overthinks a trivial fix. The knob for that is reasoning effort , the Codex counterpart of extended thinking in Claude Code, controlled by the model_reasoning_effort key. In short: set model_reasoning_effort = "high" (or another value) in ~/.codex/config.toml to change it permanently, use the -c flag for a single run, or…
Codex AI offers five levels of reasoning effort: minimal, low, medium, high, and xhigh. These levels control how much the model internally processes information before responding. To change the effort permanently, modify the model_reasoning_effort value in the ~/.codex/config.toml configuration file. For a single run, add the -c flag with the desired effort level.
During a session, you can switch reasoning effort by typing /model. Profiles allow you to save specific settings in separate configuration files for easy switching based on the task at hand. The plan_mode_reasoning_effort and agents.default_subagent_reasoning_effort settings change effort only in plan mode or for subagents. Higher reasoning effort results in slower responses but can improve accuracy, especially for complex tasks. Always confirm your model supports reasoning effort before making changes.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
