turboquant-wasm 0.2.2 → 0.2.3

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +3 -3
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -7,7 +7,7 @@
7
7
 
8
8
  WASM + relaxed SIMD build of [botirk38/turboquant](https://github.com/botirk38/turboquant) for browsers and Node.js.
9
9
 
10
- Based on the paper "TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate".
10
+ Based on the paper ["TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate"](https://arxiv.org/abs/2504.19874) (Google Research, ICLR 2026).
11
11
 
12
12
  ## What this adds
13
13
 
@@ -35,7 +35,7 @@ import { TurboQuant } from "turboquant-wasm";
35
35
 
36
36
  const tq = await TurboQuant.init({ dim: 1024, seed: 42 });
37
37
 
38
- // Compress a vector (~3 bits/dim, ~6x compression)
38
+ // Compress a vector (~4.5 bits/dim, ~6x compression)
39
39
  const compressed = tq.encode(myFloat32Array);
40
40
 
41
41
  // Decode back
@@ -86,7 +86,7 @@ Encoding preserves inner products — verified by golden-value tests and distort
86
86
  ## Credits
87
87
 
88
88
  - [botirk38/turboquant](https://github.com/botirk38/turboquant) — original Zig implementation
89
- - Google's TurboQuant paper — algorithm design
89
+ - [TurboQuant paper](https://arxiv.org/abs/2504.19874) (Google Research, ICLR 2026) — algorithm design
90
90
 
91
91
  ## License
92
92
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "turboquant-wasm",
3
- "version": "0.2.2",
3
+ "version": "0.2.3",
4
4
  "description": "TurboQuant WASM SIMD vector compression — 3 bits/dim with fast dot product. Requires relaxed SIMD (Chrome 114+, Firefox 128+, Safari 18+, Node 20+)",
5
5
  "type": "module",
6
6
  "main": "dist/index.js",