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

# Pricing

> How Levanto Sage decision calls are priced.

Sage uses **prepaid credits**. There are no seats, no subscriptions, and no minimum commitment — you pay only for the decisions you make.

## Per decision

**1 cent (USD) per decision** — each Yes/No, Choice, Scale, Sort, or Tags question in a `/decide` or `/decide/batch` call. In `/decide/batch`, every question in every group's `questions` list counts separately — shared `content` does not reduce the per-question price.

## Grounding

Optional [grounding](/decision-model/grounding) lets Sage search the web before deciding. **When a search actually runs, that decision costs an extra cent** on top of the base rate.

| What happened                                                              | You pay |
| -------------------------------------------------------------------------- | ------- |
| No grounding                                                               | \$0.01  |
| Grounding attached, but no search (`grounding_meta.triggered` is `false`)  | \$0.01  |
| Grounding attached and a search ran (`grounding_meta.triggered` is `true`) | \$0.02  |

With `trigger: low_confidence` (the default), Sage only searches when the first pass is uncertain — confident answers stay at the base rate. With `trigger: always`, every request includes the search surcharge.

<Note>
  A **402** response means your API key is valid but account balance is too low.
</Note>

## Volume discounts

Enterprise teams with high-volume workloads can get **volume discounts** on top of standard per-decision pricing.

<Card title="Talk to us" icon="envelope" href="mailto:team@levanto.ai">
  Reach out to [**team@levanto.ai**](mailto:team@levanto.ai) to discuss enterprise pricing and volume discounts.
</Card>