turbovec
A vector index built on TurboQuant, written in Rust with Python bindings
What is turbovec?
A vector index built on TurboQuant, written in Rust with Python bindings
Overview of turbovec
What it is
turbovec is an in-process vector index for Python and Rust. It stores each vector at 2 or 4 bits per dimension using Google Research's TurboQuant and searches that compressed form directly with agent-optimised SIMD kernels. According to the README, ten million 768-dimensional embeddings take 31 GB of RAM as float32, but turbovec holds them in 4 GB and searches them 4.4× faster than FAISS, at matching recall.
Key features
- Fast: 4.4× FAISS FastScan at 4-bit and 2.2× at 2-bit, on ARM and x86. A staged search scans one bit plane and rescores the shortlist exactly.
- Low memory: 7.8× smaller than float32 at 4-bit, 15.5× at 2-bit, at the same recall as FAISS.
- Online: No training step, no rebuilds; the codebook comes from the maths, not your data. Add a vector and it is searchable; remove one in microseconds.
- Incremental:
sync(path)writes only what changed since the last call — one fsync, crash-safe at any byte, milliseconds however large the index. - Local: A library, not a service; nothing leaves your process. Pair it with an open-source embedding model for a fully offline stack.
- Filtered: Pass an allowlist of ids or a bitmask; the kernel skips what the filter excludes and returns up to
kresults from the allowed set.
Use cases
turbovec is suitable for applications requiring fast approximate nearest neighbor search with low memory footprint, online updates, incremental persistence, and local processing. It is used in Python and Rust projects, and integrates with frameworks like LangChain, LlamaIndex, Haystack, and Agno for retrieval-augmented generation (RAG) pipelines. It is not intended for graph-based indexing, exact float results, server deployments, or vectors with dimensions not multiples of 8 or wider than 16,384.
Getting started
pip install turbovec
For Rust:
cargo add turbovec
See the repository README for detailed usage examples.
License and pricing
This project is licensed under the MIT License and is open source (free).
Based on the project's public README and repository metadata.