> For the complete documentation index, see [llms.txt](https://docs.xynq.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.xynq.ai/how-the-network-works.md).

# How the Network Works

This page is the high-level mental model. Deeper mechanics live in the **Architecture** pages.

### 1. You send a prompt

You type into the XYNQ terminal and hit send. Your browser packages the conversation and forwards it to the nearest **coordination endpoint**.

### 2. The scheduler builds a plan

The scheduler looks at:

* which **model** you selected,
* how that model is **sharded** across the fleet right now,
* which nodes are **online, idle, and low-latency** to you,

and assembles an execution plan — essentially a pipeline of nodes that, together, hold every layer of the chosen model.

### 3. Inference runs across the mesh

Your tokens flow through the pipeline. Each node runs its assigned layers and passes the intermediate activations to the next node. The final node produces output tokens, which are streamed straight back to your terminal as they are generated.

### 4. Idle GPUs do the work

The nodes executing your request are contributors' machines that were sitting idle. The moment a contributor starts using their computer, their node drains gracefully and the scheduler reroutes around it — your chat is unaffected.

### 5. Nothing is kept

Once your response is delivered, the working state is discarded. No transcript is written to disk anywhere in the network. Refreshing your browser erases the local copy too.

### The flywheel

```
more contributors  ->  more capacity  ->  faster & more free usage
       ^                                              |
       |                                              v
   $XYNQ rewards  <-  fees fund infra  <-  more users & demand
```

The network gets better the more people participate — both as users and as contributors.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.xynq.ai/how-the-network-works.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.