Community

Developer Docs & SDK

Relevance AI Documentation

Sign Up

Log In

Home

The low-code platform for building LLM-powered Agents and Tools.

Introduction

Get started

Agent

Tools

Data sources to be saved on Relevance and used by AI for further analysis

Data

Ready-to-use analysis tool/flow powered by LLMs.

Templates

API keys

Agent quickstart

Getting started with Agents

Tools are powerful components that allow you to create custom automations and workflows powered by Large Language Models (LLMs). They provide a flexible, low-code environment for building AI-driven solutions tailored to your specific needs.

Tools quickstart

Getting Started with Tools

Relevance AI is SOC 2 Type II compliant and works with trusted vendors.

Security overview

Plans & Credits

Plans and Credits

System quotas

Introduction to AI-powered Agents and how they can transform your workflow

Introduction to Agents

Agent Templates are pre-built AI assistants designed and maintained by Relevance. These templates provide a quick and easy way to get started with Agents for various use cases and domains.

Agent Templates

Create an Agent

How to make an API call to have a conversation with an Agent

Conversation

Conversation with an Agent via and API call

How to make an API call to list conversations with an Agent

List conversations

API calls to list the conversations with an agent

Frequently Asked Questions

Low-code builder of LLM chains and automations

Introduction to Tools

Ready-to-use Tools designed and maintained by Relevance

Tool Templates

Getting started with creating my first Tool

Create a Tool

Create a Tool - Step 1

To pass data or instruction on how to analyze the data into a Tool

User inputs - Get started

Text input

Long text input

Drop-down

Options drop down

Numeric input

Text list

JSON

JSON list

List of JSON

Input files and extract the texts from them

File to text

Temporarily save your input file on the cloud

File to URL

Table

Checkbox

What are chains?

Hold up! What are “chains”?

Analysis steps applied to the provided data

Tool steps - Get started

PDF to Text

Convert PDF to text

Extract Website Content

Knowledge search

Audio/Video to Text

Convert audio/video to text

Insert to dataset

Insert data into a dataset

Split large pieces of text to smaller chunks

Split text

Vectorize the data and perform vector search

Vectorize and search

Vectorize and Search array

Convert data in spreadsheet to their corresponding JSON format

Spreadsheet to JSON

Convert spreadsheet to JSON

Convert a JSON string to its corresponding JSON object

String to JSON

Convert string to JSON

JSON filtering

Filter JSON by fields

Combine entries in an array of string into one string

Combine string array

Combine array of strings

Code - Java

Code - JavaScript

Export data to a temporary CSV file that can later be downloaded

Export to downloadable

Export data to a temporary downloadable file

LLMs can be enhanced with your own context and data

Knowledge

Add Knowledge to LLMs

How to apply a Tool to an individual data entry

Single run

Tool execution on individual data entries

How to apply a Tool to all entries in a dataset

Bulk run

Tool execution on the whole data

Export Results

Export analysis results

Running a step only if conditions are met

Conditions

Add conditions to steps

Loops

Add loops to steps

Configure Tools outputs and get them ready for sharing with your team

Output configuration

Configure Tools Output

How to convert a string into other formats

String conversion

Convert string to another format

How to activate a Tool and allow long run time

Long run time

Long running Tools

Setting up Azure OpenAI

Common Errors

Errors

Frequently asked questions about building Tools

How to use Relevance to store and access your datasets

Introduction to Data

A single data table and the available features

Data table

Data Table

Frequently asked questions about data and data preparation

Ask GPT to find the answer to your question in a file

GPT on my files

Ask GPT to find the answer to your question in a web-page

GPT on my website

Use LLMs to retrieve the best matching answer from knowledge sources

Q&A on data

Question-Answering on data

Extract data from PDF

Categorize/Classify the text into pre-specified categories/classes

Categorize text

Summarize a meeting transcript based on the specified agenda and focus points

Summarize transcript

Summarize meeting transcript

Suggest Categories/Tags/Labels/Code-frame by looking at a subset of the data

Category/Tag suggester

Extract categories in data

Analyze the overall sentiment of an input text and label it with Positive/Negative or Neutral.

Sentiment analysis

Mask/hide personally identifiable information (PII) in text

Anonymize text

Use few-shot prompting to teach your LLM to mimic your style

Sales response automation

Teach LLMs to mimic your style

Adding support for generating Zoom meeting links

Zoom

Adding support for Redis VSS in a Relevance AI chain

Redis VSS

Ship LLM-powered features into production

Quick start

We make it super easy to build LLM-powered features in your product

defineChain

Building chains

The setup function is where you compose your chain

setup

The setup function

Define the inputs that your chain should receive

params

The params property

Library of transformations that can be used in chain steps

Step options

Instantly deploy your chains to hosted API endpoints

Deploying chains

Our data-science inspired notebook lets you build chains no-code

Notebook UI

In this guide we will create a ChatGPT-like question & answer widget for your PDFs.

ChatGPT on your PDFs

Building a ChatGPT like Q&A on PDFs

In this guide we will create an agent that analyses your SQL schema and creates a dashboard full of charts.

Get started

Agents

Tools

Data

Use cases

Integrations

SDK

Datasets

Home

Your starting point

Key concepts

Get started

Agents

Tools

Data

Use cases

Integrations

SDK

Datasets

​Your starting point

​Key concepts

Your starting point

Key concepts