GianfrancoCorrea / gpt-4-vision-chat

GPT 4 Turbo Vision with Chainlit

Welcome to GPT 4 turbo vision! 🚀🤖

Upload an image 🔗

• option 1) Drag & Drop
• option 2) Click in the "UPLOAD FILES" button, on the left of the chat input 💬
• option 3) Copy a image and paste it in the chat input (ctrl + v)

GPT-4-1106-preview for messages that ARE NOT images 📝

• change log:
  • Changed GPT-4-1106-preview for gpt-3.5-turbo-1106, due high cost of GPT-4-1106-preview

gpt-4-vision-preview for messages that ARE images 📷

If you upload more than 1 image, it will take the first image, this is just for demo purposes

• change log:
  • Added Low res mode, to test the fidelity of responses
  • Changed max_tokens from 300 to 100, to test the costs.
  • Changed max_tokens from the output to 300
  • Cleared image history after the response
  • Image size limit set up to 1mb

⚠ There is a limit of 100 requests per day for gpt-4-vision-preview

If a limit reached error appears, you can wait 15 minutes and try again or in a few hours. It is an openai limit.

install dependencies (recomended to use a virtualenv)

pip install -r requirements.txt

add the api key to the environment (.env)

OPENAI_API_KEY=<your api key>

run the app

chainlit run app.py

run the app with a specific port

chainlit run app.py --port 5001