Skip to content

Latest commit

 

History

History
63 lines (46 loc) · 2.32 KB

File metadata and controls

63 lines (46 loc) · 2.32 KB

File Attachments

← Back to Main README | Features | Installation | Parameters | Web Search | API Reference | Contributing


IntelliLLM Playground supports uploading files alongside your prompts to enhance your interactions with LLMs that have multimodal capabilities.

Supported File Types

  • Images:
    • PNG
    • JPEG
    • WebP
  • Documents:
    • PDF

Usage

  1. Attaching Files:

    • Click the image or document icon in the prompt editor toolbar
    • Select a file from your device
    • The file will be attached to your prompt and displayed below the editor
  2. Managing Attachments:

    • You can remove attached files by clicking the "X" button next to the file name
    • Files are included in API requests to the LLM
    • The UI clearly indicates which files are attached to each prompt
  3. Using Attachments in Prompts:

    • Reference attached images or documents in your prompt text
    • For example: "Analyze the image I've attached and describe what you see."
    • The LLM will receive both your text prompt and the attached files

Technical Implementation

  • Images: Sent as base64-encoded data in the messages array with type image_url
  • Documents: Included with type file in the appropriate format
  • Size Limits:
    • Images: Maximum size of 10MB
    • Documents: Maximum size of 25MB

Best Practices

  1. Image Quality:

    • Use clear, high-quality images for best results
    • Crop images to focus on relevant content
    • Consider image resolution and file size
  2. Document Handling:

    • PDFs work best when they contain searchable text rather than scanned images
    • For multi-page documents, consider referencing specific pages in your prompt
  3. Prompt Design:

    • Be specific about what you want the LLM to focus on in attached files
    • For images, specify if you want detailed descriptions, analysis of specific elements, or other outputs
    • For documents, clarify if you want summaries, specific information extraction, or analysis

Limitations

  • File attachments do not persist between sessions
  • Some models may have limitations on how they process certain file types
  • Very large or complex files may not be processed effectively by all models