Tune Assistant in Action

The Assistant offers a robust framework for developers to create AI assistants tailored to their applications.

For more detailed instructions and advanced concepts, please refer to the Assistant page under Concepts.

​

Getting Started

To begin working with AI assistants, you have two options:

1. Interactive Assistant Playground: Prototype and test Assistant functionality quickly without writing code.
2. Integrate with the Assistant API:: Build Assistant functionality directly into your applications.

​

Option 1: Create Assistant using Playground

Follow these simple steps to create your first assistant via playground:

1. Build an Assistant by Providing a Model ID: This ID will be used in all APIs.
2. Set Specific Instructions: Define the instructions that the assistant needs to follow. Defines your assistant’s behavior.
3. Choose the Response Model: Select a model that combines responses from multiple tools to generate the final output.
4. Select the Function Calling Model: Pick a model that enables the assistant to decide which function to call.
5. Pre-Set Tools: Equip your assistant with tools such as:
   • Search: Enable the assistant to perform web searches (note: it cannot visit the search results).
   • Image Generation: Allow the assistant to generate creative images.
6. Specify Actions: Define the actions your assistant can perform. Actions serve as messengers between your LLMs and external applications, helping you:
   • Interact with External Apps: Connect with outside applications using RESTful APIs by providing an OpenAPI spec.
   • Retrieve Data: Fetch information from external applications, such as recent orders.
   • Take Actions: Perform tasks in other applications, like sending an email.
7. Function Calling: Describe your function to the model. It allows you to access real-time data and execute custom functions within a specific context.

​

Option 2: Create Assistant using API

This section focuses on using the Assistants API to create and manage assistants. You can view complete assistant API reference doc here.

Prerequisites

• Tune API key: Get it from View API Keys under your profile.
• Organization ID: Get it under your Organization settings or click on “Get Code” button inside playground

​

Create Basic Assistant

To create a new assistant, send a POST request to the /v1/assistants endpoint.

Required Fields:

• modelName: A unique name for your assistant within your organization.
• functionModel: Specifies the model used for function calling.
• responseModel: Specifies the model used for generating responses.

Required Headers:

• x-tune-key: Your Tune API key.
• X-Org-Id: Your organization ID. If not provided, the default organization is used. You can find your organization ID here.

Here’s an example of creating a basic assistant:

Replace MODEL_ID with the ID of the models you wish to use.

curl -X POST "https://studio.tune.app/v1/assistants" \
 -H 'accept: application/json' \
 -H 'content-type: application/json' \
 -H 'x-tune-key: <YOUR-TUNE-KEY>' \
 -H 'X-Org-Id: <YOUR-ORG-ID>' \
 -d '{
  "modelName": "my-first-assistant",
  "functionModel": {
    "modelId": "MODEL_ID"
  },
  "responseModel": {
    "modelId": "MODEL_ID"
  }
}'

​

Managing Assistants

After creating an assistant, you might want to list all your assistants, retrieve details of a specific assistant, or delete an assistant. Here’s how you can do that using the Assistants API.

​

Listing Assistants

To fetch a list of your assistants, send a GET request to the /v1/assistants endpoint.

Example Request:

curl -X GET "https://studio.tune.app/v1/assistants" \
 -H "accept: application/json" \
 -H "x-tune-key: <YOUR-TUNE-KEY>" \
 -H "X-Org-Id: <YOUR-ORG-ID>"

Optional Query Parameters:

• limit: Maximum number of assistants to return (default is 10).
• order: Order of results (asc or desc).
• after: Pagination token for fetching the next page.
• before: Pagination token for fetching the previous page.

Example with Query Parameters:

curl -X GET "https://studio.tune.app/v1/assistants?limit=5&order=asc" \
 -H "accept: application/json" \
 -H "x-tune-key: <YOUR-TUNE-KEY>" \
 -H "X-Org-Id: <YOUR-ORG-ID>"

​

Getting an Assistant

To retrieve details of a specific assistant, send a GET request to the /v1/assistants/{assistantId} endpoint, replacing {assistantId} with the ID of your assistant.

Example Request:

curl -X GET "https://studio.tune.app/v1/assistants/<YOUR-ASSISTANT-ID>" \
 -H "accept: application/json" \
 -H "x-tune-key: <YOUR-TUNE-KEY>" \
 -H "X-Org-Id: <YOUR-ORG-ID>"

This request will return details such as the assistant’s modelName, functionModel, responseModel, and any associated tools.

​

Deleting an Assistant

To delete an assistant, send a DELETE request to the /v1/assistants/{assistantId} endpoint.

Example Request:

curl -X DELETE "https://studio.tune.app/v1/assistants/<YOUR-ASSISTANT-ID>" \
 -H "accept: application/json" \
 -H "x-tune-key: <YOUR-TUNE-KEY>" \
 -H "X-Org-Id: <YOUR-ORG-ID>"

For more detailed guide for assistant checkout the Assistant concept page.

​

Next Steps

Now that you’ve created and managed a basic assistant, you can explore more advanced features and details, checkout:

• Complete Assistant Guide - Deep dive into assistant workflow, models, tools, and configurations
• Modifying Assistant: Update your assistant’s configuration as needed.
• Custom Tools Guide - Create your own functions and integrations. And using built-in tools to enhance capabilities with tools like web search, image generation etc.
• Actions Guide - Integrate external APIs and services.
• Assistant API Reference - Complete API documentation.