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

# AI chat

> Chat with any LLM, your connected knowledge, and the web — all in one place.

Use Hadiq.io Chat to ask questions, analyze data, generate content, and get answers grounded in your organization's knowledge. Connect your preferred LLM, upload files on the fly, and use built-in tools like web search, code execution, and image generation.

## Choose your model

Hadiq.io connects to all major LLM providers. Select a model from the model picker in any chat session.

<CardGroup cols={2}>
  <Card title="OpenAI" icon="circle">
    GPT-4o, o1, o3-mini, and other OpenAI models
  </Card>

  <Card title="Anthropic" icon="robot">
    Claude Sonnet, Claude Opus, and Haiku variants
  </Card>

  <Card title="Google" icon="google">
    Gemini 2.0 Flash, Gemini Pro, and other Gemini models
  </Card>

  <Card title="Self-hosted" icon="server">
    Ollama, LiteLLM, vLLM, and any OpenAI-compatible endpoint
  </Card>
</CardGroup>

Your admin configures which providers and models are available to your team. If you don't see the model you need, ask your admin to add the provider under **Admin → LLM Providers**.

## What you can do in chat

<CardGroup cols={2}>
  <Card title="Web search" icon="globe">
    Get up-to-date information from the web. Hadiq.io supports Serper, Google PSE, Brave, SearXNG, and an in-house web crawler.
  </Card>

  <Card title="File uploads" icon="file">
    Upload documents on the fly. Supported formats include PDF, DOCX, PPTX, XLSX, CSV, Markdown, images, and code files.
  </Card>

  <Card title="Code interpreter" icon="code">
    Execute code in a sandbox to analyze data, render charts, or transform files. Results appear directly in the chat.
  </Card>

  <Card title="Image generation" icon="image">
    Generate images from text prompts using supported image generation models configured by your admin.
  </Card>

  <Card title="Voice mode" icon="microphone">
    Speak your questions and hear responses using built-in speech-to-text and text-to-speech.
  </Card>

  <Card title="Artifacts" icon="sparkles">
    Generate downloadable documents, graphics, and other files as outputs from your conversation.
  </Card>
</CardGroup>

## Enable web search

To pull in live information from the web, enable **Web Search** in the toolbar before sending your message. Hadiq.io fetches relevant pages, summarizes the content, and includes citations in the response.

<Tip>
  Use web search when you need current information not in your connected knowledge sources — for example, recent news, pricing changes, or public documentation.
</Tip>

## Upload files

Drag a file into the chat input or click the attachment icon to upload. Hadiq.io processes the file and uses its content as context for your conversation.

Supported file types include:

* **Documents**: PDF, DOCX, PPTX, TXT, Markdown
* **Spreadsheets**: XLSX, CSV
* **Code**: Most common source file extensions
* **Images**: PNG, JPG, WEBP, and others (with vision-capable models)

<Note>
  File uploads are scoped to the current chat session. To make a file permanently available to an agent, attach it to the agent's knowledge sources instead.
</Note>

## Run code

Enable the **Code Interpreter** tool to let Hadiq.io write and execute code in a sandboxed environment. Use this to:

* Analyze data from an uploaded CSV or spreadsheet
* Generate charts and visualizations
* Perform calculations or data transformations
* Create or modify files

Output logs and any generated files are returned in the conversation.

## Share a chat

Share a conversation with a teammate so they can read the context without you having to copy-paste.

<Steps>
  <Step title="Open the chat session you want to share">
    Navigate to the conversation in the sidebar.
  </Step>

  <Step title="Click the Share icon">
    Find the **Share** icon in the top-right corner of the chat window.
  </Step>

  <Step title="Copy the link">
    Copy the generated link and send it to your teammate. Recipients can view the chat but cannot continue the conversation from your session.
  </Step>
</Steps>

Admins can review shared chats and monitor usage across the organization from the analytics dashboard.

## Deep Research

For complex questions that require in-depth investigation across multiple sources, use **Deep Research** instead of a standard message. Deep Research runs a multi-step agentic process and produces a structured report.

[Learn more about Deep Research →](/features/deep-research)