This article provides a comprehensive overview of API Reference via Python SDK - PyGEAI, outlining its structure and available functionality.
n
To use the API, you must authenticate each request using API Tokens. These tokens are managed in the Console.
The following properties are common to every interaction with the API Reference:
| Variable |
Description |
| GEAI_API_BASE_URL |
$BASE_URL. The base URL for your Globant Enterprise AI installation (e.g., https://api.saia.ai or a custom value). |
| GEAI_API_KEY |
$GEAI_APITOKEN. An API token generated for each project or organization. |
The API is designed with three levels of abstraction to address different use cases and preferences:
Provides a simple way to interact with Globant Enterprise AI using simple commands. It handles communication with the REST API and presents the responses in an easy-to-read format.
Offers more granular control over API interactions. You work directly with client classes that handle the HTTP requests and responses. This layer returns data in JSON format, providing flexibility for further processing.
Abstracts the complexity of handling raw JSON data. Manager classes map JSON responses to Python objects, simplifying data manipulation and integration with application logic.
Each item in this section has these three ways of implementing a function. You can choose the one that best suits your use case.
The API includes the following functionality:
- Interact with the Globant Enterprise AI chat system to generate chat completions using specified models and parameters.
- Generate embeddings from various input types, including text and images.
- Manage projects within your organization.
- Interact with the RAG Assistants in Globant Enterprise AI.
Since version 2025-05.