BigHugger
GH Repository · RyanCodrai

turbovec

A vector index built on TurboQuant, written in Rust with Python bindings

stars
17,208
30-day movement
+340113/day
Related entries
61
Connections
2
Rustturboquantquantannavx512ragembeddingsembeddingfaissnearest-neighborneonquantizationpythonrustsimdvector-search

turbovec is a vector index built on TurboQuant, implemented in Rust with Python bindings. The repository is MIT licensed and its topics indicate SIMD acceleration (AVX512, NEON), approximate nearest-neighbor search, and quantization for embedding-based vector search.

You want a Rust-based, quantized vector index usable from Python for RAG-style nearest-neighbor search.

Use it to

  • Run approximate nearest-neighbor search over embeddings
  • Use quantized vector indexing in a RAG pipeline
  • Call a Rust vector index from Python
  • Leverage SIMD (AVX512/NEON) for vector search performance

For Developers building RAG or vector search in Python

Role
rag
Language
Rust
Licence
MIT
Forks
1,471
Open issues
17
Last push
2026-09-13
Skills shipped
1
topicsvector-searchquantizationrustannragpython