Skip to content

Latest commit

 

History

History
29 lines (24 loc) · 1.82 KB

README.md

File metadata and controls

29 lines (24 loc) · 1.82 KB

UHHGPT Image Upload

Utilize the capability of GPT-4o Vision on the UHHGPT web portal

📙 Demo

Demo Gif
Image used can be found on Pixabay

🛑 Disclaimer

This is a Proof of Concept (PoC) and is not officially affiliated with the University of Hamburg.

⭐ Features

  • Allows users to upload images to the UHHGPT web portal and interact with GPT-4o for inquiries
    • Feature not yet available: Saving past image queries for extended discussions. (Current state: One query and one image. If you want to continue the conversation about the image you may upload it again. Should be fixed in the next releases)
  • QoL: Auto select GPT-4o on website load
  • QoL: Autohide usage hint banner on website load
  • Support for multiple browsers running Tampermonkey (Safari currently not supported)
    • Tested on: Google Chrome, Brave Browser and Firefox. Should work on other Chromium or Gecko based browsers as well
  • Display past uploaded images below user message

⚙️ Installation

  1. Install the Tampermonkey extension
  2. Open the Tampermonkey Dashboard
  3. Add a new script
  4. Copy contents of uhhgpt-image.user.js to the newly created script or open the raw version so Tampermonkey can prompt you to install the script directly

🍩 Acknowledgements / Credits

onbeforescriptexecute polyfill code by jspenguin2017 (MIT License)