# Paddock

> Run local models as dependable infrastructure, not a pile of processes.

**Category:** Local model operator
**Status:** Public alpha · Linux x86-64
**Role:** The local inference layer

Paddock manages llama.cpp model lifecycle, OpenAI-compatible routing, downloads, telemetry, and a live operator dashboard on one machine.

## Who it is for

Operators who want a clear, local-first control plane for GGUF models and the applications that depend on them.

## Capabilities

### Loads models on demand

Apply deterministic capacity eviction and optional idle unloading across model roles.

### Presents one local API

Route supported chat, completion, Responses, embedding, and reranking requests through an OpenAI-compatible surface.

### Owns the model lifecycle

Download, validate, register, start, stop, restart, and inspect GGUF model sets.

### Keeps operations visible

Watch model, process, request, and GPU behavior without retaining prompts or responses.

## Working sequence

1. **Install the verified alpha.** Use the public bootstrap and review the current compatibility evidence.
2. **Build the managed runtime.** Let Paddock detect the host and build its pinned llama.cpp baseline.
3. **Register a model.** Pull a GGUF artifact set and expose it through the local API when requested.

## Boundaries

- The alpha is published for Linux x86-64; binary availability is not broad hardware certification.
- Paddock binds to loopback by default and does not provide an application authentication boundary.
- llama.cpp is the production backend. vLLM support is experimental and operator-supplied.

## Next steps

- [Get Paddock](https://get.kinra.ai/paddock/)
- [Read the docs](/docs/paddock/)
- [Documentation](https://kinra.ai/docs/paddock/)

_Source authority: The Paddock repository and the release publication at get.kinra.ai/paddock._