@bergetai/pi-provider
Run [Berget AI](https://berget.ai) models inside [Pi](https://pi.dev).
Package details
Install @bergetai/pi-provider from npm and Pi will load the resources declared by the package manifest.
$ pi install npm:@bergetai/pi-provider- Package
@bergetai/pi-provider- Version
0.1.16- Published
- May 29, 2026
- Downloads
- 572/mo · 22/wk
- Author
- bergetbjork
- License
- MIT
- Types
- extension
- Size
- 25.3 KB
- Dependencies
- 0 dependencies · 2 peers
Pi manifest JSON
{
"extensions": [
"./index.ts"
]
}Security note
Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.
README
Pi extension for Berget AI
Run Berget AI models inside Pi.
Prerequisites
Installation
pi install npm:@bergetai/pi-provider
Restart Pi or run /reload to load the extension.
Authentication
This provider supports two authentication methods. API key authentication takes precedence when both are configured.
Using an API key
Set your API key using one of the following methods.
Environment variable (recommended):
export BERGET_API_KEY=your-api-key
~/.pi/agent/auth.json:
{
"berget": {
"type": "api_key",
"key": "your-api-key"
}
}
~/.pi/agent/models.json:
{
"providers": {
"berget": {
"baseUrl": "https://api.berget.ai/v1",
"apiKey": "your-api-key"
}
}
}
Using a Berget Code plan
If you have a Berget Code plan, authenticate interactively through your browser:
/login
Select "Use a subscription" and then "Berget AI".
Usage
Once authenticated, Berget AI models are available in Pi. Select one with:
/model
Or cycle through models with Ctrl+P. Models are prefixed with berget/.
List all available models from the command line:
pi --list-models
Contributing
See CONTRIBUTING.md for development setup, testing, and architecture details.
Licence
MIT