Adding Custom Providers

Beyond the 23 built-in providers, you can add custom providers through the plugin system.

Plugin Providers

Plugins can register new AI providers that appear alongside built-in ones in the Settings UI. This is useful for:

• Proprietary or internal LLM endpoints
• Custom model hosting (e.g., vLLM, TGI)
• Specialized embedding or image generation services
• Proxy services with custom authentication

Implementation

A plugin exposes providers through its providers export. Each provider entry declares a definition (the implementation), display metadata, and its capabilities so KinBot knows where to use it.

// In your plugin's main file
export const providers = {
  'my-llm': {
    definition: myProviderImplementation,
    displayName: 'My Custom LLM',
    capabilities: ['llm', 'embedding'],
    noApiKey: false,
    apiKeyUrl: 'https://my-service.com/keys',
  },
}

KinBot automatically registers plugin providers with a plugin_<name>_ prefix (e.g., plugin_my-plugin_my-llm). Once registered, the provider appears in Settings > Providers alongside built-in ones and can be configured with an API key and base URL.

OpenAI-Compatible Endpoints

Many self-hosted solutions expose an OpenAI-compatible API. For these, you can often use the built-in OpenAI or OpenRouter provider with a custom base URL, without needing a plugin:

1. Go to Settings > Providers > OpenAI
2. Set the Base URL to your endpoint (e.g., http://localhost:8000/v1)
3. Set the API key if required

This works with vLLM, llama.cpp server, LocalAI, and other OpenAI-compatible services.

Ollama

For local models, Ollama is the recommended approach. It’s a built-in provider that requires no API key:

1. Install and run Ollama on your machine
2. Pull models: ollama pull llama3 or ollama pull nomic-embed-text
3. In KinBot, configure the Ollama provider with the base URL (default: http://localhost:11434)

From Docker, use http://host.docker.internal:11434 instead of localhost.