# Chat

Chat is the heart of i10x.ai. It looks like a simple chat window - but underneath it is one of the most powerful AI environments available. This page explains every feature so you can get the most out of every conversation.

{% hint style="info" %}
💡 Just getting started? Leave everything on default settings and type your first question. The Auto model will handle the rest.
{% endhint %}

### Choosing Your AI Model

At the bottom left of the chat bar you will see the model selector. By default it is set to Auto.

Auto means i10x.ai analyzes your prompt and automatically routes it to the best available model for that specific task. You do not need to know the difference between GPT-5 and Gemini to get great results.

Want more control? Click the model selector to manually choose a model from the list below.

| Model                             | Best For                                           | Speed       |
| --------------------------------- | -------------------------------------------------- | ----------- |
| **Auto**                          | Everything - let i10x decide                       | ⚡ Fast      |
| **GPT-5**                         | Complex reasoning, long documents, nuanced writing | 🐢 Thorough |
| **GPT-5 Mini**                    | Quick everyday tasks, summaries, Q\&A              | ⚡ Fast      |
| **GPT-5 Nano**                    | Ultra-fast simple tasks                            | ⚡⚡ Fastest  |
| G**PT-4.1**                       | Reliable all-rounder, writing and coding           | ⚡ Fast      |
| **Gemini 2.5 Pro**                | Massive documents, multi-modal reasoning           | 🐢 Thorough |
| **Gemini 2.5 Flash**              | Fast and efficient everyday tasks                  | ⚡ Fast      |
| **Gemini 3.1 Pro Preview**        | Advanced reasoning and research                    | 🐢 Thorough |
| **Gemini 3.1 Flash Lite Preview** | Speed-optimised tasks                              | ⚡⚡ Fastest  |
| **Gemini 3 Pro Preview**          | Balanced reasoning and speed                       | ⚡ Fast      |
| **Claude 4.6 Sonnet**             | Writing, analysis, nuanced conversation            | ⚡ Fast      |
| **Claude 4.6 Opus**               | Deep reasoning and complex analysis                | 🐢 Thorough |
| **Claude Haiku 4.5**              | Fast and lightweight tasks                         | ⚡⚡ Fastest  |
| **Grok 4**                        | Real-time web knowledge, current events            | ⚡ Fast      |
| **Grok 4.1 Fast**                 | Speed-first tasks                                  | ⚡⚡ Fastest  |
| **Grok Code Fast**                | Code generation and debugging                      | ⚡ Fast      |
| **DeepSeek V3**                   | Technical reasoning and STEM tasks                 | ⚡ Fast      |
| **Meta LLaMA 4 Scout**            | Lightweight open-source tasks                      | ⚡ Fast      |
| **Meta LLaMA 4 Maverick**         | More powerful open-source tasks                    | 🐢 Thorough |
| **Z.ai**                          | Experimental and creative tasks                    | ⚡ Fast      |

{% hint style="info" %}
💡 Not sure which model to pick? Here is the simple rule:\
Use Auto for everything. Switch to GPT-5 or Claude 4.6 Opus when you need deep analysis. Switch to GPT-5 Nano, Gemini Flash or Claude Haiku when you need speed.
{% endhint %}

### Prompt Optimizer

Not getting the result you expected? Click the wand icon (✨) next to the input bar before sending your message.

i10x will automatically rewrite and improve your prompt to extract the best possible response from the AI - no prompt engineering experience needed.

Think of it as having a professional prompt engineer on standby, for free.

### Voice Input

Prefer to talk rather than type? Click the microphone icon in the chat bar to speak your prompt. i10x will transcribe your voice and send it to the AI instantly.

{% hint style="info" %}
⚠️ Note: Voice input is supported. Voice playback (AI speaking back to you) is not available yet.
{% endhint %}

### The "+" Menu - Supercharge Your Chat

Click the blue + icon in the chat bar to unlock four powerful features that give the AI more context about you and your work.

<details>

<summary>🧠 <strong>Memory</strong></summary>

Memory is active by default. This means i10x remembers important context from your past conversations - your preferences, your projects, your name - so you never have to repeat yourself.

You can toggle Memory OFF if you want a completely fresh, anonymous session.

</details>

<details>

<summary>📎 <strong>Upload Files</strong></summary>

Attach documents directly to your chat for instant analysis. Supported formats include:

PDF, Word (.docx), Excel (.xlsx), Google Docs exports, and more.

Best for one-off tasks:\
→ "Summarize this contract."\
→ "Find the key numbers in this spreadsheet."\
→ "What are the action items in this meeting transcript?"

Need the AI to remember a document permanently across multiple chats? Use Knowledge Base instead.

</details>

<details>

<summary>🌐 <strong>Web Search</strong></summary>

Enable Web Search to give the AI access to live, real-time information from the internet - including breaking news, current prices, recent research papers, and more.

Turn this ON whenever your question involves anything time-sensitive or recent.

Example:\
→ "What are the top AI funding deals this week?"\
→ "What is the current Bitcoin price?"\
→ "Summarize the latest research on GLP-1 drugs."

</details>

<details>

<summary><strong>📚 Knowledge Base</strong></summary>

Your Knowledge Base is a permanent document library connected to your i10x account. Create folders (e.g. "Client Contracts," "Research Papers," "Company SOPs") and the AI can reference them in any future chat.

The difference between File Upload and Knowledge Base:\
→ File Upload = one session, one document, gone when the chat ends.\
→ Knowledge Base = permanent, always available, works across all chats.

To set up your Knowledge Base, click "Knowledge Base" in the left sidebar of your i10x dashboard.

</details>

### Connect Your Apps

Click the Connector icon (🔌) in the chat bar to link your favourite apps directly to your chat. Once connected, you can talk to your real data - no copy-pasting, no switching tabs.

| App                 | What You Can Ask                                                 |
| ------------------- | ---------------------------------------------------------------- |
| **Gmail**           | "Summarize my last 5 unread emails and flag anything urgent."    |
| **Google Calendar** | "What does my Tuesday look like? Find me a free 30-minute slot." |
| **Google Drive**    | "Find the Q3 report I uploaded last week and summarize it."      |
| **Google Docs**     | "Expand the introduction of my proposal document."               |
| **Google Sheets**   | "Analyze the data in my sales tracker and identify trends."      |
| **Google Slides**   | "Summarize the key points from my pitch deck."                   |
| **Google Forms**    | "What are the most common responses in my feedback form?"        |
| **Google Tasks**    | "What tasks do I have due this week?"                            |
| **Notion**          | "Update my Q3 roadmap page with these meeting notes."            |
| **Firefly**         | "Access and reference my creative assets."                       |

{% hint style="info" %}
✅ Try This Now:

1. Click the Connector icon (🔌) and connect Gmail.
2. Make sure Web Search is ON.
3. Type: "Summarize my last 5 unread emails and flag anything urgent."

You just built your first AI email assistant. Welcome to i10x.ai.
{% endhint %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://knowledge.i10x.ai/chat-and-research/chat.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
