Google â
Gemini â
This guide introduces how to configure and use the Google Gemini model in Aide.
You can find more detailed information in the Google Gemini Official Reference Documentation.
API Base URL Configuration â
You need to configure aide.openaiBaseUrl
to https://{LOCATION}-aiplatform.googleapis.com/v1beta1/projects/{PROJECT}/locations/{LOCATION}/endpoints/openapi
Note
Location
is the geographic location of your Google Cloud project, such asus-central1
.Project
is the name of your Google Cloud project.
For example, if your Google Cloud project name is my-project
and the geographic location is us-central1
, then your API Base URL would be:
https://us-central1-aiplatform.googleapis.com/v1beta1/projects/my-project/locations/us-central1/endpoints/openapi
.
API Key Configuration â
You need to configure aide.openaiKey
as your Google Gemini API Key.
For detailed steps to obtain the API Key, please refer to the official reference documentation mentioned above.
Model Configuration â
You need to configure aide.openaiModel
to the Google Gemini model. We recommend using the google/gemini-1.5-flash-001
model. For more models, please refer to the official reference documentation mentioned above.
Example Configuration File â
Below is a complete configuration example:
{
"aide.openaiBaseUrl": "https://{LOCATION}-aiplatform.googleapis.com/v1beta1/projects/{PROJECT}/locations/{LOCATION}/endpoints/openapi",
"aide.openaiKey": "your-google-gemini-api-key",
"aide.openaiModel": "google/gemini-1.5-flash-001"
}
Make sure to replace "your-google-gemini-api-key"
with your actual API Key, and fill in your specific location information and project name in {LOCATION}
and {PROJECT}
.