Turbovec – Google's TurboQuant for vector search in Rust

esafak

lancedb and duckdb integrations would be great...

nharada

It would be nice to have the README be a little more human written for a project where you actually want people to adopt it

ghm2199

Wow! 4GB for 10 million documents. This means one could build a reverse index much faster than before and devx processes like debugging, performance testing would become much smoother. Can't wait for the sqlite bindings to come out!

burgerboii

Who is this co-author called t <t@t>?

zuzululu

what could i use this for as part of my agentic workflow? codebase indexing? docs ?

anishvarghese

This looks perfect for local, privacy first search, but since it's built in Rust, has anyone tried compiling it to WASM to run directly inside a browser extension?

sp1982

If anyone is looking to retrofit to an existing pipeline, I use similar ideas to compress vectors for job search, getting roughly 8x compression with about a 3.5% drop in quality. My experiment: https://corvi.careers/blog/vector-search-embedding-compressi...

spoaceman7777

Well. That is insane. O_O Fantastic job!

refulgentis

Bloviating nonsense, 3rd time I’ve seen something like this in HN since TurboQuant came out. You don’t need float32, never did. Source: I’ve been writing on device embedding code for 4 years.

beernet

Why not just use Qdrant? They've been integrating TurboQuant for months, works well.

cute_boi

Another vibe coded slop where they can't even spend time on Readme or documentation around code...

Eridrus

FAISS is no longer close to SoTA: https://ann-benchmarks.com/index.html https://vector-index-bench.github.io/ https://big-ann-benchmarks.com/neurips23.html

bobmarleybiceps

people should read turboquant's open review comments: https://openreview.net/forum?id=tO3ASKZlok

cat-whisperer

What's a good embedding model and search to run locally? something fast and lightweight.

OutOfHere

I am not convinced that Turbovec yields better retrieval than the same amount of bits of a Matryoshka embedding.

lmeyerov

Interestingly, while we don't fine-tune generative models for Louie.ai, we found fine-tuning embedding models to be a major $ saver. Instead of 1K-2K wide frontier embedding vector lens... Just 64. Huge savings on vector DB $$$. I'm curious how that works with something like turboquant. Not needed any more, still dominant, better together, ... .