> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mnrl.app/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Chat: Conversations with intelligent assistants

> Use Monorail AI Chat to hold human-like conversations with AI assistants. Get ideas, draft content, research topics, and more in real time.

AI Chat gives you access to intelligent, human-like assistants powered by models from OpenAI, Anthropic, xAI, and Cohere. Use Chat to brainstorm ideas, get answers, draft content, or simply explore what AI can do.

## Getting started with Chat

<Steps>
  <Step title="Open your dashboard and select Chat">
    Navigate to your Monorail AI dashboard and click **Chat** in the sidebar.
  </Step>

  <Step title="Choose an AI assistant">
    Select from the available AI assistants powered by models from OpenAI, Anthropic, xAI, and Cohere.
  </Step>

  <Step title="Type your message or question">
    Enter your message or question in the input field at the bottom of the screen.
  </Step>

  <Step title="Send your message">
    Press Enter or click **Send** to receive a response from the assistant.
  </Step>

  <Step title="Continue the conversation">
    Keep the conversation going — Chat remembers context within the session, so you can build on previous messages without repeating yourself.
  </Step>
</Steps>

## What you can do with Chat

* Brainstorm ideas and creative concepts
* Research topics and get summarized answers
* Draft, edit, or improve written content
* Answer questions on any subject
* Get help with tasks, planning, or analysis

## Supported capabilities

<Note>
  Chat supports file analysis, real-time data access, web browsing, and AI YouTube analysis on all paid plans.
</Note>

## Credit usage

Chat costs credits based on the AI model you select and the number of tokens in your conversation — both input and output count toward your total. As a reference, GPT-4o costs approximately 1 credit per 2,800 input tokens. Longer conversations accumulate more tokens over time and therefore cost more.

See the [Credits](/credits) page for a full breakdown of model rates.