zanllp/infinite-image-browsing

What it solves

Infinite Image Browsing (IIB) solves the problem of managing and navigating massive collections of AI-generated images and videos. It replaces basic file explorers with a high-performance browser that can index thousands of files, extract generation metadata (prompts, models, LoRAs), and provide advanced search and organization tools.

How it works

IIB indexes image folders and generates thumbnails for fast browsing. It parses metadata from various AI software (like Stable Diffusion, ComfyUI, and Fooocus) to create searchable tags. For advanced organization, it uses AI embeddings via OpenAI-compatible APIs to perform semantic clustering, automatically grouping similar images based on prompt meaning and generating descriptive folder names using an LLM.

Who it’s for

AI artists and prompt engineers who generate large volumes of media and need a way to efficiently search, compare, and organize their outputs across different generation tools.

Highlights

• Broad Compatibility: Supports Stable Diffusion web UI, ComfyUI, Fooocus, NovelAI, StableSwarmUI, and Invoke.AI.
• AI-Powered Organization: Features "Smart Organize" for semantic clustering and natural-language retrieval of images based on prompt similarity.
• High Performance: Uses caching and thumbnail pre-generation to display thousands of images in milliseconds.
• Advanced Navigation: Includes a "Walk Mode" to flatten folder structures for seamless browsing and a TikTok-style vertical view.
• Metadata Integration: Full-screen preview with nested JSON navigation for EXIF and generation data.
• Flexible Deployment: Available as an SD-webui extension, a standalone Python program, or a desktop application.