Google Gemini Integration
The Google plugin for Enthusiast, makes it possible to use Gemini models.
Install the Plugin
First, install the enthusiast-model-google
package using pip:
pip install enthusiast-model-google
Generate API key
- Navigate to https://aistudio.google.com/apikey
- Log in.
- Hit "Create API key" button in upper right corner.
Set environment variable
GOOGLE_API_KEY=<your_gemini_api_key>
Enable the relevant language model and embedding providers in the settings.py
:
server/pecl/settings.py
CATALOG_LANGUAGE_MODEL_PROVIDERS = {
"Google": "enthusiast_model_google.GoogleLanguageModelProvider",
}
CATALOG_EMBEDDING_PROVIDERS = {
"Google": "enthusiast_model_google.GoogleEmbeddingProvider",
}
Save the changes and restart the web server and the worker.
Configuring a Data Set to use Google Gemini
After installing and configuring the plugin, you can now select Google Gemini as a language model and embedding provider when setting up a new data set.