# CrabDash Local Models

Run ~700 MLX community models locally on Apple Silicon — no Docker, no Ollama, no separate process.

CrabDash can route to local models alongside cloud providers. Browse ~700 models from the [MLX Community](https://huggingface.co/mlx-community) catalog, pull them from the menubar, and serve them through the same `127.0.0.1:5635` endpoint as cloud models. Your app doesn't need to know the difference.

## [MLX (Apple Silicon native)](/content/docs/crabdash/local-models#mlx-apple-silicon-native/index.html)

On Apple Silicon Macs, CrabDash runs models directly via MLX — no separate process, no Docker, no Ollama required.

### [Pull a model](/content/docs/crabdash/local-models#pull-a-model/index.html)

1. Click the menubar icon  
2. Open **Models → Local**  
3. Browse the MLX catalog (~700 models) or search by name  
4. Click **Pull** — the model downloads to `~/.crabdash/models/`

### [Route to a local model](/content/docs/crabdash/local-models#route-to-a-local-model/index.html)

Once pulled, the model appears in `/v1/models`. Use it like any other model:

```
curl http://127.0.0.1:5635/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{\n    "model": "mlx-community/Llama-3.2-3B-Instruct-4bit",\n    "messages": [{"role": "user", "content": "Hello"}]\n  }'
```

Memory requirement: roughly 1 GB per billion parameters at 4-bit quantization. A 16 GB Mac can comfortably run 7B models alongside other apps.

## [Routing between local and cloud](/content/docs/crabdash/local-models#routing-between-local-and-cloud/index.html)

You can route specific models or apps to local models while keeping others on cloud providers. In Settings → Routing:

- **By model**: Pin a model name to the local provider  
- **By app**: Route requests from a specific app (identified by API key) to local-only

This is useful for keeping development traffic local (free, private) while using cloud models for production workloads.

### CrabDash Gateway API

Use CrabDash's local OpenAI-compatible and Anthropic-compatible endpoints from any app, SDK, or CLI tool.