TurboQuant-WASM
AI
vector search
SHOW HN2026
How HN took it
1 positive, 1 neutral or asking, 0 mixed, 2 critical, out of 4 comments about the project.
Performance: 0 for, 1 againstIdea: 0 for, 1 againstDesign: 1 for, 0 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Browser search demo and use cases look interesting
From 2 comments
Awesome! Also love the gaussian splat demo, cool use case!
hhthrowaway1230 on HNVery cool. I added the new multi embedding 2 model to my site the other week from google I guess need to dig into this and see if it’s faster and has more use cases! Thanks for…
glohbalrob on HN
Objections
Quantized search is slower than unquantized float32 search
From 2 comments
I tried TQ for vector search and my findings is not good, it is not worth it if you cannot use GPU, however I got same quality of search as 32f using 8bit quant I wrote ann ext…
netdur on HNSloppiest slop I've seen in a couple weeks: - fork of a fork of a quantization technique - Only contribution is...compiling JS to WASM by default? - suspicious burst of ~nothing…
refulgentis on HN
Saves memory but makes text embedding search much slower in the demo
From 1 comment
Sloppiest slop I've seen in a couple weeks: - fork of a fork of a quantization technique - Only contribution is...compiling JS to WASM by default? - suspicious burst of ~nothing…
refulgentis on HN
Only a fork of an existing quantization technique with little original contribution
From 1 comment
Sloppiest slop I've seen in a couple weeks: - fork of a fork of a quantization technique - Only contribution is...compiling JS to WASM by default? - suspicious burst of ~nothing…
refulgentis on HN
Requests
Benchmark speed improvements in the demo
From 1 comment
Sloppiest slop I've seen in a couple weeks: - fork of a fork of a quantization technique - Only contribution is...compiling JS to WASM by default? - suspicious burst of ~nothing…
refulgentis on HN
Questions
Whether OPQ-style trained quantization would perform better
From 1 comment
So i assumed it would get crushed by OPQ (which requires training)
ninja3925 on HN
Alternatives named
OPQ (optimized product quantization)
From 1 comment
So i assumed it would get crushed by OPQ (which requires training)
ninja3925 on HN