Discord bot built with Pycord that offers AI chat and image generation inside of Discord. Image generations are saved in cloudinary and hosted online for all to download.
This PR integrates OpenAI's powerful vision AI model into the discord bot to enable image recognition capabilities.
The vision model can analyze images sent in discord channels and identify objects, scenes, faces etc with a high degree of accuracy. It provides labels, confidence scores and bounding boxes on detected items.
Some key highlights:
Add new /vision command to run the vision model on images
Vision model returns structured JSON output with labels, scores and bounding boxes
Cache image embeddings to improve performance and reduce compute costs
Set up API keys and authentication to call OpenAI API
Output vision results as an embed in discord
The vision model opens up many new possibilities like automatically tagging images, detecting inappropriate content, searching images by content etc. We can build more commands in the future based on the vision capabilities.
Add OpenAI Vision Model
This PR integrates OpenAI's powerful vision AI model into the discord bot to enable image recognition capabilities.
The vision model can analyze images sent in discord channels and identify objects, scenes, faces etc with a high degree of accuracy. It provides labels, confidence scores and bounding boxes on detected items.
Some key highlights:
The vision model opens up many new possibilities like automatically tagging images, detecting inappropriate content, searching images by content etc. We can build more commands in the future based on the vision capabilities.
Let me know if any changes are needed!