RyanCodrai/turbovec: A vector index built on TurboQuant, written in Rust with Python bindings
Summary
turbovec implements Google Research’s TurboQuant as a local Rust vector index with Python bindings.
Quantize without a corpus-training phase.
A shared random rotation makes coordinate distributions predictable enough for precomputed Lloyd-Max codebooks, with optional per-coordinate calibration for finite-dimensional drift.
Score compressed vectors directly.
ARM and x86 kernels use lookup tables and SIMD instructions instead of reconstructing every database vector before comparison.
Keep mutation online.
New vectors enter the index immediately, stable external IDs support constant-time removal, and the corpus can grow without rebuilding a trained quantizer.
A 10 million document corpus takes 31 GB of RAM as float32. turbovec fits it in 4 GB - and searches it faster than FAISS.
Add vectors, they're indexed — no train step, no parameter tuning, no rebuilds as the corpus grows.
Push filters into the kernel.
Candidate allowlists skip empty 32-vector blocks before lookup and discard individual excluded slots before heap insertion.
Filtering happens inside the SIMD kernel at 32-vector block granularity
Publish the measurement boundary.
The repository exposes result files and reproducible suites, explains low-dimensional recall limits, and labels its fast synthetic harness as an optimization screen rather than evidence for published claims.
It is a screening tool, not a source of published numbers.
Read the original ↗Save to your Sponge
Summarized with Sponge. Quotes are the author’s own words.