# Opencode

## install

```bash
curl -fsSL https://opencode.ai/install | bash
```

## configure custom llama.cpp server

Opencode supports any OpenAI-compatible API. Set the following environment variables to point it at your llama.cpp server:

```bash
export OPENAI_API_KEY=""
export OPENAI_BASE_URL="http://driveripper.reeselink.com:8000/v1"
```

### persist across sessions

Add the exports to your shell profile (`~/.bashrc`, `~/.zshrc`, etc.):

```bash
echo 'export OPENAI_API_KEY=""' >> ~/.bashrc
echo 'export OPENAI_BASE_URL="http://driveripper.reeselink.com:8000/v1"' >> ~/.bashrc
source ~/.bashrc
```

### pick a model

After configuring the environment, launch opencode and select the model available from your llama.cpp instance:

```bash
opencode
```

Inside opencode, use `/model` to list available models and switch between them.

### verify the connection

Run this one-liner to confirm opencode can reach the server:

```bash
OPENAI_API_KEY="" OPENAI_BASE_URL="http://driveripper.reeselink.com:8000/v1" opencode --help
```

If no auth-related errors appear, the endpoint is reachable.