CLIP Interrogator analyzes an uploaded image using OpenAI's CLIP model and outputs a text prompt describing its style, medium, subject, and composition — effectively reverse-engineering the kind of prompt that could have produced a similar image in a text-to-image model like Stable Diffusion. Pick from a 'best', 'classic', or 'fast' analysis mode depending on how much detail you want.

It's a staple utility for the Stable Diffusion and broader generative-image community: designers and prompt engineers use it to figure out how to describe a reference photo or piece of art in prompt form, or to understand what specific keywords are driving a particular visual style. Because it's open source, it's mirrored across dozens of independent Hugging Face Spaces beyond the original by creator pharmapsychotic.

The tool is completely free to run — Hugging Face Spaces host it without a paywall, though performance depends on the compute tier of whichever mirror you use. There's no account, subscription, or usage limit tied to the core tool itself.