Type what you're in the mood for, in plain words, and it finds manga that match by meaning, not just keywords. It's our working demo: the same machine, pointed at your catalog, helps customers find products, parts, or documents.
Under the hood: React + Vite front end, FastAPI backend, Qdrant vector database, OpenAI text-embedding-3-small, deployed on GCP (Cloud Run and Firebase Hosting).
Semantic search over vector embeddings. Proof the machine works.
Visit komikfind → Data powered by MangaDex.