s-emanuilov/litepali
LitePali is a minimal, efficient implementation of ColPali for image retrieval and indexing, optimized for cloud deployment.
Implements late-interaction matching with multi-vector representations from the ColPali architecture, enabling efficient semantic search across document images without requiring traditional PDF parsing. Decouples image processing from PDF extraction by working exclusively with image inputs, reducing dependencies and allowing CPU-based PDF conversion in separate pipelines. Provides deterministic batch indexing and index persistence for reproducible retrieval workflows in resource-constrained cloud environments.
122 stars. No commits in the last 6 months. Available on PyPI.
Stars
122
Forks
11
Language
Python
License
MIT
Category
Last pushed
Oct 07, 2024
Commits (30d)
0
Dependencies
1
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/s-emanuilov/litepali"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
unum-cloud/UForm
Pocket-Sized Multimodal AI for content understanding and generation across multilingual texts,...
rom1504/clip-retrieval
Easily compute clip embeddings and build a clip retrieval system with them
mazzzystar/Queryable
Run OpenAI's CLIP and Apple's MobileCLIP model on iOS to search photos.
Ubaida-M-Yusuf/Makimus-AI
AI-powered media search — find images and videos using natural language or visual queries
slavabarkov/tidy
Offline semantic Text-to-Image and Image-to-Image search on Android powered by quantized...