RAG
5 projects
turbovec fits 10M vectors in 4 GB and outruns FAISS
RyanCodrai/turbovec
A Rust index built on Google Research's TurboQuant: no training step, hand-written NEON and AVX-512 kernels, crash-safe incremental saves and filtered search.
Skill Seekers turns docs sites and PDFs into agent skills
yusufkaraaslan/Skill_Seekers
Eighteen source types in, twenty-two targets out, with automatic conflict detection and an AI scan that emits one config per framework it finds in your project.
Zvec is a vector database that lives inside your process
alibaba/zvec
Alibaba's embedded C++ engine does dense, sparse, full-text and hybrid search with WAL durability — SQLite's deployment story, applied to retrieval.
Hindsight is agent memory that tries to learn, not recall
vectorize-io/hindsight
A memory system claiming state of the art on LongMemEval, with the rare detail that outside researchers reproduced the score instead of the vendor asserting it.
PixelRAG retrieves over screenshots instead of parsed text
StarTrail-org/PixelRAG
A Berkeley project that renders pages as images and searches them visually, keeping the tables, charts and layout that HTML parsing throws away.