Build and deploy AI-driven assistants with our OpenAI Assistants Template. This tutorial provides a hands-on approach to using OpenAI's Assistant API, complete with code modules, interactive Jupyter Notebook examples, and best practices to get you started on creating intelligent conversational agents.
Welcome to the OpenAI Assistants Template! This repository is a step-by-step tutorial 📚 for leveraging OpenAI's powerful Assistant API to build intelligent and conversational AI assistants. Whether you're a developer, a student, or just an AI enthusiast, this guide will help you harness the power of GPT models for your projects.
To get the most out of this tutorial, you should have:
example.env
to .env
and insert your OpenAI API key.pip install -r requirements.txt
to install dependencies.Explore the OpenAI-Assistant-Template.ipynb
for a hands-on experience that walks you through the capabilities of the Assistant API, enriched with practical examples.
Dive into modules.py
to find utility functions and classes that provide a cleaner and more maintainable codebase, making it easier to build upon.
modules.py
🧰create_assistant(client, name, description, instructions, tools=[], model="")
: Create a new Assistantget_assistant(client, assistant_id)
: Retrieve an existing assistant using an assistant ID.start_new_chat(client)
: Start a new conversation with the AI assistant.get_chat(client, thread_id)
: Retrieve an existing conversation using a thread ID.add_message(client, thread, content)
: Send a new message to the Assistant within a conversation thread.get_messages_in_chat(client, thread)
: Fetch all the previous messages within a conversation thread.run_chat(client, thread, assistant)
: Process the conversation thread through the assistant for generating responses.Your contributions make the open-source community an incredible arena for innovation. If you've got ideas on how to make this template even better, your pull requests are welcome! Let's make learning AI with OpenAI an exciting journey for everyone.
This project is open-sourced under the MIT License. Feel free to use it as you wish.