Assistants API

# Assistants API

[**Assistants API Beta**](https://platform.openai.com/docs/assistants/overview/agents)

The Assistants API allows you to build AI assistants within your own applications. An Assistant has instructions and can leverage models, tools, and knowledge to respond to user queries. The Assistants API currently supports three types of [tools](https://platform.openai.com/docs/assistants/tools): Code Interpreter, Retrieval, and Function calling. In the future, we plan to release more OpenAI-built tools, and allow you to provide your own tools on our platform.

Explore the capabilities of the Assistants API using the [Assistants playground](https://platform.openai.com/playground?mode=assistant) or by building a step-by-step integration outlined in this guide. A typical integration of the Assistants API includes:

1. **Create an Assistant:** Define custom instructions and pick a model for your [Assistant](https://platform.openai.com/docs/api-reference/assistants/createAssistant). Enable tools like Code Interpreter, Retrieval, and Function calling.
2. **Create a Thread:** Start a [Thread](https://platform.openai.com/docs/api-reference/threads) when a user begins a conversation.
3. **Add Messages:** Include [Messages](https://platform.openai.com/docs/api-reference/messages) in the Thread as users ask questions.
4. **Run the Assistant:** Trigger responses by [running](https://platform.openai.com/docs/api-reference/runs) the Assistant on the Thread, automatically calling the relevant tools.

The Assistants API is in **beta**. We welcome your feedback in our [Developer Forum](https://community.openai.com/)!

This starter guide details the key steps to create and run an Assistant using the [Code Interpreter](https://platform.openai.com/docs/assistants/tools/code-interpreter).

### Step 1: Create an Assistant

[**Create an Assistant**](https://platform.openai.com/docs/assistants/overview/step-1-create-an-assistant)

Configure an Assistant to respond to Messages using parameters such as:

- **Instructions:** Define how the Assistant and model should behave or respond.
- **Model:** Choose from GPT-3.5 or GPT-4 models, including fine-tuned ones. The Retrieval tool requires `gpt-3.5-turbo-1106` and `gpt-4-1106-preview` models.