/xai-google-gemini

Xircuits components for Google's Gemini model.

Primary LanguagePythonMIT LicenseMIT

Google Generative AI Components for Xircuits

This repository offers a suite of components designed to integrate the Google Generative AI library into the Xircuits framework. These components facilitate various operations, including content generation, chat sessions, and token counting, leveraging Google's advanced AI models.

Prerequisites

  • Python 3.8 or higher

Installation

Once Xircuits is installed, simply run

xircuits install google_gemini

You may also manually clone this repository to xai_components and install the requirements.txt.

Usage

It is recommended to store your Google API key in an environment variable for security and ease of use, or export it via:

export GOOGLE_API_KEY='your_api_key_here'

Components Overview

The library includes the following components for seamless integration with Google's Generative AI capabilities:

  • GeminiAuthorize: Handles authorization with the Google Generative AI library using an API key, which can be provided directly or fetched from an environment variable named GOOGLE_API_KEY.
  • GeminiGenerateContent: Generates content based on provided input data using a specified generative model.
  • GeminiGenerateContentStream: Offers real-time content generation through streaming, similar to GeminiGenerateContent.
  • GeminiStartChat: Initiates a chat session using a specified generative model.
  • GeminiChatSendMessage: Sends a message and receives a response within an active chat session.
  • GeminiChatSendMessageStream: Enables real-time chat interaction through streaming, similar to GeminiChatSendMessage.
  • GeminiCountTokens: Counts the number of tokens in the provided input data using a specified generative model.