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

# Pricing

> Pay only for what we save you. No savings, no fee.

## How pricing works

OpenCompress uses a **pay-for-savings** model. We charge **20% of the money we save you**. If compression doesn't save you anything on a particular request, there's no additional fee.

```
Your cost = LLM cost (on compressed tokens) + 20% × savings
```

## Example

Say you send a 10,000-token prompt that we compress to 6,000 tokens using GPT-4o (\$2.50/1M input tokens):

|                        | Tokens       | Cost               |
| ---------------------- | ------------ | ------------------ |
| Without OpenCompress   | 10,000 input | \$0.0250           |
| With OpenCompress      | 6,000 input  | \$0.0150           |
| **Savings**            | 4,000 tokens | **\$0.0100**       |
| OpenCompress fee (20%) |              | \$0.0020           |
| **Your total cost**    |              | **\$0.0170**       |
| **Net savings**        |              | **\$0.0080 (32%)** |

You save **32%** on this call. We earn \$0.002. Everyone wins.

## Router vs BYOK pricing

<CardGroup cols={2}>
  <Card title="Router Mode" icon="route">
    **You pay us:** LLM cost + 20% fee

    We handle the LLM provider. Simple, no setup.
  </Card>

  <Card title="BYOK Mode" icon="key">
    **You pay us:** 20% fee only

    You pay your LLM provider directly. We only charge the compression fee.
  </Card>
</CardGroup>

## Billing details

* **Minimum deposit:** \$10
* **Deposit options:** $10, $50, $100, $500, \$1,000, or custom amount
* **Payment:** Stripe (credit card)
* **Balance:** Prepaid, deducted per request
* **Refunds:** Available for unused balance

## Output savings

When we compress your input, the model's output also tends to be shorter. We estimate output savings conservatively:

```
output_savings = input_compression_rate × 0.5
```

This means a 40% input compression results in an estimated 20% shorter output. Both input and output savings are factored into your total savings calculation.