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

# How Pareto works

> GLM 5.3 Flash, served on Pareto GPUs.

Pareto serves GLM 5.3 Flash on its own GPUs. You pay for the tokens you use.

## Connect your tools

Add your Pareto endpoint and API key to a router that supports custom OpenAI-compatible providers. Your router sends Flash requests to Pareto. See the [integration guides](/integrations/routers).

You can also [call the API directly](/sdk-examples).

## Served by Pareto

Pareto processes these requests on its own GPUs.

When the account has no credits for a request, or too many requests arrive at the same time, the API returns **HTTP 429**. When Pareto is at capacity, or a temporary failure occurs, the API returns **HTTP 503**. When a model request fails, the API returns **HTTP 502**. Your router can retry or use another provider if you configure it to do so.

See [Troubleshooting](/errors#429-too-many-requests) for retry guidance.

## Serving system details

We do not disclose the technical details of our GLM 5.3 Flash serving system. This helps prevent competitors from reverse-engineering our stack.

## Pay per token

Add a card and buy prepaid credits. Input, cached input, and output tokens use your balance at the [published rates](/pricing).

## Data handling

Pareto stores request content to operate, secure, and improve the service. We do not use this content to train or evaluate AI models. We do not sell your data to anyone. Pareto does not offer zero data retention (ZDR).

See the [Privacy Policy](https://paretoinference.com/privacy) for details on what Pareto collects, how we use it, and how long we keep it. If you use a router, its data policy also applies to the requests it handles.