> 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/node-lifecycle.md).

# Node Lifecycle

A node moves through a well-defined lifecycle. The client manages all of it automatically.

### States

| State                   | Meaning                                                                                    |
| ----------------------- | ------------------------------------------------------------------------------------------ |
| **Provisioning**        | Client installed; node registering with the coordinator.                                   |
| **Syncing**             | Downloading and loading assigned model shards into VRAM.                                   |
| **Online / Idle-armed** | Ready; will serve work when the GPU is idle.                                               |
| **Serving**             | Actively executing inference slices.                                                       |
| **Draining**            | Owner reclaimed the GPU (or node is leaving); finishing in-flight work, refusing new work. |
| **Offline**             | Disconnected; replicas cover its shards.                                                   |

### Transitions

```
Provisioning -> Syncing -> Online ⇄ Serving
                     \           |
                      \          v
                       ---->  Draining -> Offline
```

* **Online ⇄ Serving** toggles with demand and your idle policy.
* **Draining** is entered the moment you start using your machine; it's graceful and fast.
* **Offline** nodes are automatically covered by other replicas, so the network never stalls.

### Rebalancing

When your node joins or leaves, the coordinator may **rebalance** shards across the fleet to maintain replication targets and locality. This is invisible to users and handled in the background.

### Health & reputation

Throughout its life, a node accrues a **reputation score** from uptime, correctness on verification checks, and responsiveness. Higher reputation → more work routed → more rewards.


---

# 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/node-lifecycle.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.