> ## Documentation Index
> Fetch the complete documentation index at: https://docs.narrative.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Thinking Budget Too Large

> The thinking budget leaves no room for an answer.

**HTTP status:** 400

## Error message

```
thinking budget {budget_tokens} must be smaller than max_tokens {max_tokens}, or there is no room for an answer
```

## When this error occurs

`thinking.budget_tokens` is at or above `max_tokens`. Thinking tokens are spent from the same output
budget as the answer, so a budget that fills `max_tokens` leaves the model nothing to answer with —
the turn gets truncated mid-thought and the run fails later, with nothing useful to show for the
spend.

## How to fix

Raise `max_tokens`, lower `budget_tokens`, or both. Leave clear headroom: the budget is a ceiling on
thinking alone, and the answer still has to fit in what remains.

| `max_tokens` | Reasonable `budget_tokens` |
| ------------ | -------------------------- |
| 2048         | 1024 (the floor)           |
| 8000         | up to \~4096               |
| 32000        | up to \~16000              |