Every industry today relies on visual inspection, from manufacturing quality checks to infrastructure monitoring yet manual comparisons are slow, inconsistent, and prone to human error. That’s what inspired us to build V8 - Visual Difference Engine, an AI-powered visual inspection system that identifies and describes image changes using LLM. We wanted to create something that could see differences the human eye might miss, but without the heavy lifting of training custom machine learning models. Instead, we used cutting-edge LLM to analyze and interpret visual changes with semantic understanding not just pixel differences. Drop your images, and V8 does the rest detecting every change in seconds.

How We Built It Frontend: A clean, minimal UI where users can upload images. Backend: A Python FastAPI server handles uploads and sends them to the LLM for analysis. Image Processing: The LLM analyzes image patterns using computer vision and CNNs to detect visual differences, then returns detailed reasoning with visual highlights.

What We Learned How to leverage LLMs for image analysis through multi-image prompts.

Challenges We Faced Managing API rate limits and latency for multiple high-resolution images.

Built With

Share this project:

Updates