> ## Documentation Index
> Fetch the complete documentation index at: https://reedai-07fa30f1.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Files & Images

> Upload, analyze, and work with files and images in your conversations

## File Upload Overview

ZeroTwo supports uploading various file types directly into your conversations. AI can read, analyze, and work with your files to provide contextual assistance.

<Note>
  Files are processed securely and only used within your conversation context. They are not used to train AI models.
</Note>

## Supported File Types

ZeroTwo supports a wide range of file formats for different use cases.

<Tabs>
  <Tab title="Documents">
    **Text Documents**:

    * `.txt` - Plain text files
    * `.md` / `.mdx` - Markdown documents
    * `.pdf` - PDF documents
    * `.doc` / `.docx` - Microsoft Word
    * `.rtf` - Rich text format

    **Spreadsheets**:

    * `.csv` - Comma-separated values
    * `.xlsx` / `.xls` - Microsoft Excel
    * `.tsv` - Tab-separated values

    **Presentations**:

    * `.ppt` / `.pptx` - Microsoft PowerPoint
  </Tab>

  <Tab title="Code & Development">
    **Source Code**:

    * `.js` / `.jsx` / `.ts` / `.tsx` - JavaScript/TypeScript
    * `.py` - Python
    * `.java` - Java
    * `.cpp` / `.c` / `.h` - C/C++
    * `.cs` - C#
    * `.go` - Go
    * `.rs` - Rust
    * `.rb` - Ruby
    * `.php` - PHP
    * `.swift` - Swift
    * `.kt` - Kotlin

    **Configuration**:

    * `.json` - JSON data
    * `.yaml` / `.yml` - YAML configuration
    * `.xml` - XML data
    * `.toml` - TOML configuration
    * `.env` - Environment variables

    **Web Development**:

    * `.html` - HTML documents
    * `.css` / `.scss` / `.sass` - Stylesheets
    * `.vue` - Vue components
    * `.svelte` - Svelte components
  </Tab>

  <Tab title="Images">
    **Raster Images**:

    * `.jpg` / `.jpeg` - JPEG images
    * `.png` - PNG images
    * `.gif` - GIF images (including animated)
    * `.webp` - WebP images
    * `.bmp` - Bitmap images
    * `.ico` - Icon files

    **Vector Graphics**:

    * `.svg` - Scalable vector graphics

    <Info>
      Vision-capable models (like GPT-4 Vision, Claude 3, Gemini Pro Vision) can analyze image content, describe scenes, extract text, and answer questions about images.
    </Info>
  </Tab>

  <Tab title="Data & Specialized">
    **Database & Data**:

    * `.sql` - SQL scripts
    * `.db` / `.sqlite` - SQLite databases
    * `.jsonl` - JSON Lines

    **Archives**:

    * `.zip` - ZIP archives
    * `.tar` / `.gz` - TAR/GZIP archives

    **Logs & Outputs**:

    * `.log` - Log files
    * `.out` - Output files

    **Other**:

    * `.ipynb` - Jupyter notebooks
    * `.sh` / `.bash` - Shell scripts
    * `.dockerfile` - Docker files
  </Tab>
</Tabs>

## Uploading Files

There are multiple ways to add files to your conversation.

<Steps>
  <Step title="Drag and drop">
    Drag files directly from your file manager into the chat input area.

    <Tip>
      You can drag multiple files at once to upload them together.
    </Tip>
  </Step>

  <Step title="Click to upload">
    Click the **📎 Attach** button in the chat input, then select files from your file picker.

    **Keyboard shortcut**: `Cmd/Ctrl + U`
  </Step>

  <Step title="Paste from clipboard">
    Copy an image to your clipboard and paste it directly into the chat input.

    **Keyboard shortcut**: `Cmd/Ctrl + V`

    <Info>
      This works great for screenshots! Use your system's screenshot tool, then paste directly.
    </Info>
  </Step>
</Steps>

## File Size Limits

<CardGroup cols={2}>
  <Card title="Free Plan" icon="user">
    * **Max file size**: 25 MB
    * **Max files per message**: 10 files
    * **Total upload limit**: 1 GB per month
  </Card>

  <Card title="Pro Plan" icon="star">
    * **Max file size**: 100 MB
    * **Max files per message**: 20 files
    * **Total upload limit**: 10 GB per month
  </Card>

  <Card title="Team Plan" icon="users">
    * **Max file size**: 200 MB
    * **Max files per message**: 50 files
    * **Total upload limit**: Unlimited
  </Card>

  <Card title="Enterprise Plan" icon="building">
    * **Max file size**: 500 MB
    * **Max files per message**: Unlimited
    * **Total upload limit**: Unlimited
  </Card>
</CardGroup>

<Warning>
  Files larger than the limit will be rejected with an error message. Consider compressing large files or splitting them into smaller parts.
</Warning>

## Working with Images

AI can analyze and understand images using vision-capable models.

### Image Analysis Capabilities

**What AI can do with images**:

<AccordionGroup>
  <Accordion title="Describe and identify">
    * Describe scenes, objects, and people
    * Identify logos, brands, and text
    * Recognize patterns and designs
    * Detect colors and composition

    **Example prompt**:

    ```
    What objects are in this image?
    ```
  </Accordion>

  <Accordion title="Extract text (OCR)">
    * Read text from images and screenshots
    * Extract data from forms and documents
    * Parse handwritten notes (with varying accuracy)
    * Read text from photos of signs, menus, etc.

    **Example prompt**:

    ```
    Extract all text from this image and format it as a table
    ```
  </Accordion>

  <Accordion title="Analyze diagrams and charts">
    * Interpret flowcharts and diagrams
    * Read and explain data from charts
    * Understand technical drawings
    * Analyze architecture and design

    **Example prompt**:

    ```
    Explain what this diagram shows and describe the workflow
    ```
  </Accordion>

  <Accordion title="Code from screenshots">
    * Convert code screenshots to text
    * Debug code from error screenshots
    * Explain code shown in images
    * Recreate UI from mockups

    **Example prompt**:

    ```
    Convert this code screenshot to editable text and fix any errors
    ```
  </Accordion>

  <Accordion title="Compare images">
    * Find differences between images
    * Compare versions of designs
    * Identify changes over time
    * Match similar images

    **Example prompt**:

    ```
    What are the differences between these two images?
    ```

    <Tip>
      Upload multiple images in one message to compare them.
    </Tip>
  </Accordion>
</AccordionGroup>

### Selecting Vision Models

Not all models support image analysis. Use these vision-capable models:

<CardGroup cols={2}>
  <Card title="GPT-4o" icon="eye">
    Fast, accurate vision with strong OCR
  </Card>

  <Card title="Claude 3.5 Sonnet" icon="eye">
    Excellent detail analysis and reasoning
  </Card>

  <Card title="Gemini 2.0 Flash" icon="eye">
    Fast vision with good accuracy
  </Card>

  <Card title="Gemini 1.5 Pro" icon="eye">
    Deep analysis for complex images
  </Card>
</CardGroup>

<Info>
  Switch to a vision-capable model using the model selector before uploading images for analysis.
</Info>

## Working with Documents

### Reading Documents

AI can read and analyze document content:

**Document capabilities**:

* Extract key information and summaries
* Answer questions about document content
* Find specific information
* Compare multiple documents
* Generate reports from data

**Example prompts**:

```
Summarize the main points from this PDF
```

```
What are the key findings in this research paper?
```

```
Extract all email addresses and phone numbers from this document
```

### Code Files

Upload code files for analysis, debugging, and improvements:

**Code file capabilities**:

* Code review and suggestions
* Bug detection and fixes
* Documentation generation
* Code refactoring
* Adding comments and explanations
* Converting between languages

**Example prompts**:

```
Review this code and suggest improvements
```

```
Add comprehensive comments to this Python file
```

```
Find potential security issues in this code
```

<Tip>
  Upload multiple related files (like a full project) to get context-aware suggestions.
</Tip>

### Spreadsheets and Data

Upload CSV, Excel, and data files for analysis:

**Data capabilities**:

* Statistical analysis
* Data visualization recommendations
* Finding patterns and trends
* Data cleaning and formatting
* Creating summaries and reports
* Answering questions about the data

**Example prompts**:

```
Analyze this sales data and identify trends
```

```
Create a Python script to visualize this CSV data
```

```
Find outliers in this dataset
```

## Managing Uploaded Files

### Viewing Attached Files

Files attached to messages appear as cards in the chat:

**File card information**:

* 📎 File icon and name
* 📊 File size
* 🗑️ Remove button (before sending)

<Info>
  After sending a message, files become part of that message and cannot be individually removed. You can delete the entire message if needed.
</Info>

### File Context

Files remain in context for the current conversation:

**Context behavior**:

* Files are available for reference throughout the chat
* AI can refer back to previously uploaded files
* File content is included in the conversation context
* Large files may affect response time due to processing

<Warning>
  Very large files or many files may consume significant context. For long conversations, consider starting a new chat if performance degrades.
</Warning>

## Files Page

Access all your uploaded files from the **Files** page in the sidebar.

### File Management Features

**Files page capabilities**:

* 📁 View all uploaded files
* 🔍 Search files by name or type
* 🗂️ Filter by file type
* 📅 Sort by upload date
* 🔗 See which conversations used each file
* 🗑️ Delete files you no longer need

<Steps>
  <Step title="Open Files page">
    Click **Files** in the sidebar navigation
  </Step>

  <Step title="Browse or search">
    Scroll through your files or use the search bar to find specific files
  </Step>

  <Step title="View details">
    Click a file to see:

    * File preview (for supported types)
    * Upload date and size
    * Conversations where it was used
    * Download option
  </Step>

  <Step title="Manage files">
    * Click **Download** to save a copy
    * Click **Delete** to remove the file
    * Click a conversation link to jump to where the file was used
  </Step>
</Steps>

## Best Practices

<AccordionGroup>
  <Accordion title="Optimize file sizes">
    **For better performance**:

    * Compress images before uploading
    * Use PDF instead of large Word documents
    * Split very large files into smaller parts
    * Remove unnecessary data from spreadsheets

    **Tools to compress**:

    * Images: Use WebP format or reduce resolution
    * PDFs: Use PDF compression tools
    * Archives: Use maximum compression settings
  </Accordion>

  <Accordion title="Provide context">
    **Help AI understand your files**:

    * Explain what the file contains
    * Specify what you want to know
    * Mention important sections or data
    * Clarify any unusual formatting

    **Good example**:

    ```
    I've uploaded a sales report from Q3. Can you analyze the revenue trends 
    in the "Monthly Sales" tab and compare them to the targets in column F?
    ```

    **Poor example**:

    ```
    What does this say?
    ```
  </Accordion>

  <Accordion title="Use appropriate models">
    **Choose the right model**:

    * **Images**: Use vision models (GPT-4o, Claude 3.5, Gemini)
    * **Code**: Use code-specialized models (GPT-4, Claude 3.5)
    * **Large documents**: Use models with large context windows
    * **Data analysis**: Models good at reasoning (o1, Claude 3.5)

    <Info>
      See [Models & Providers](/overview/models-and-providers) for detailed model capabilities.
    </Info>
  </Accordion>

  <Accordion title="Privacy considerations">
    **Protect sensitive information**:

    * Don't upload files with passwords or API keys
    * Redact personal information from documents
    * Be cautious with confidential business data
    * Review files before uploading

    **Remember**:

    * Files are encrypted in transit and at rest
    * Files are not used to train AI models
    * You can delete files anytime from the Files page
    * Enterprise plans offer additional security controls
  </Accordion>
</AccordionGroup>

## File Upload Troubleshooting

<AccordionGroup>
  <Accordion title="Upload fails or times out">
    **Possible causes**:

    * File exceeds size limit for your plan
    * Network connection issues
    * Unsupported file type
    * Server temporarily unavailable

    **Solutions**:

    1. Check file size against your plan limits
    2. Verify file type is supported
    3. Try uploading again
    4. Compress the file if too large
    5. Check your internet connection
  </Accordion>

  <Accordion title="Image not being analyzed">
    **Check these items**:

    * ✅ Using a vision-capable model
    * ✅ Image format is supported
    * ✅ Image is not corrupted
    * ✅ Asked specific questions about the image

    **Switch to a vision model**:

    1. Click model selector in the header
    2. Choose GPT-4o, Claude 3.5, or Gemini
    3. Upload image again
    4. Ask your question
  </Accordion>

  <Accordion title="File content not understood">
    **Improve AI understanding**:

    * Provide more context about the file
    * Specify what you're looking for
    * Ask more specific questions
    * Break complex requests into steps
    * Mention file structure or format

    **Example**:

    ```
    This CSV has sales data. Column A is dates, Column B is revenue. 
    What was the average revenue in March?
    ```
  </Accordion>

  <Accordion title="Cannot see uploaded files">
    **Troubleshooting steps**:

    1. Refresh the page
    2. Check if upload completed successfully
    3. Look for error messages during upload
    4. Visit the Files page to see all uploads
    5. Clear browser cache if files don't appear
  </Accordion>
</AccordionGroup>

## Examples

### Document Analysis

```
I've uploaded our quarterly report. Please:
1. Summarize the key financial metrics
2. Identify the top 3 growth areas
3. Note any concerning trends
```

### Code Review

```
Review this React component I've uploaded and suggest:
- Performance optimizations
- Better error handling
- Accessibility improvements
- Code style issues
```

### Image to Code

```
I've attached a screenshot of a UI design. Can you create the HTML 
and CSS to recreate this layout? Use flexbox for the main layout.
```

### Data Visualization

```
Analyze this sales data CSV and:
1. Calculate monthly growth rates
2. Identify the best-performing products
3. Create a Python script with matplotlib to visualize the trends
```

### Multi-file Project

```
I've uploaded 5 files from my Node.js project. Please review the 
architecture and suggest how to better organize the code for scalability.
```

## Next Steps

<CardGroup cols={2}>
  <Card title="Edit & Regenerate" icon="arrows-rotate" href="/chat/edit-and-regenerate">
    Learn to refine AI responses
  </Card>

  <Card title="Custom Instructions" icon="sliders" href="/chat/custom-instructions">
    Set default file handling preferences
  </Card>

  <Card title="Canvas" icon="pen-to-square" href="/tools/canvas">
    Create and edit documents collaboratively
  </Card>

  <Card title="Code Interpreter" icon="code" href="/tools/code-interpreter">
    Execute code and analyze data interactively
  </Card>
</CardGroup>

<Check>
  Upload files to give AI the context it needs to provide accurate, relevant assistance!
</Check>