# Extended Thinking

> A model mode that allocates extra inference tokens to deliberate before answering, often with a configurable thinking budget.

- Category: Reasoning & Inference
- Stage: Rising
- Also called: Thinking mode, Visible chain-of-thought budget
- Canonical: https://www.joelcsaji.com/ai-dictionary/extended-thinking

## Plain English

You pay for the model to think harder before it talks.

## Example

A hard debugging prompt runs with a higher thinking budget so the model explores alternatives before proposing a fix.

## Why it matters

Thinking budgets trade latency and cost for better multi-step reasoning on hard tasks.

## Related

- [Reasoning Model](https://www.joelcsaji.com/ai-dictionary/reasoning-model)
- [Test-Time Compute](https://www.joelcsaji.com/ai-dictionary/test-time-compute)
- [Reasoning Effort](https://www.joelcsaji.com/ai-dictionary/reasoning-effort)

## Sources

- [Anthropic: Extended Thinking](https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking)
- [OpenAI: Reasoning Models](https://developers.openai.com/api/docs/guides/reasoning)