# alwaysThinkingEnabled

> The key that can switch off thinking in all your chats.

**Field:** Artificial Intelligence  
**Category:** Claude Code  
**Also known as:** alwaysThinkingEnabled setting

## Think of it like

It is like a switch that turns off thinking out loud. Leaving it on changes nothing.

## Definition

alwaysThinkingEnabled is a settings key. Set to false, it turns off extended thinking in all sessions. Thinking is on by default, so true does nothing. Some models always think, such as Opus 5.5. They ignore false. The MAX_THINKING_TOKENS variable wins over it for one session.

## Example

Emma uses an old model for quick fixes and wants fast replies. She sets alwaysThinkingEnabled to false.

## Example code

```json
{
  "alwaysThinkingEnabled": false
}
```

Turn thinking off where the model allows it.

## Why it matters

Thinking costs time and tokens, and some quick tasks do not need it.

## Common mistake

Setting it to true to turn thinking on. It is already on, so true does nothing.

## Three principles

- **False turns it off.** Every session.
- **True does nothing.** Already on.
- **Some models ignore it.** They always think.

## Related terms

- [MAX_THINKING_TOKENS](https://learn.benibix.com/term/ai/max-thinking-tokens-variable)
- [CLAUDE_CODE_DISABLE_THINKING](https://learn.benibix.com/term/ai/claude-code-disable-thinking-variable)
- [effortLevel](https://learn.benibix.com/term/ai/effort-level-setting)

---

Source: https://learn.benibix.com/term/ai/always-thinking-enabled-setting
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-17
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
