Vector (and Scalar) Quantization, in Pytorch

🔴 MiniSom is a minimalistic implementation of the Self Organizing Maps

[CVPR 2025] 🔥 Official impl. of "TokenFlow: Unified Image Tokenizer for Multimodal Understanding and Generation".

ECCV 2022, Oral, VQFR: Blind Face Restoration with Vector-Quantized Dictionary and Parallel Decoder

[ICCV 2025] SimVQ: Addressing Representation Collapse in Vector Quantized Models with One Linear Layer

⚡ A fast embedded library for approximate nearest neighbor search

[ICLR 2025] Binary Spherical Quantization + [CVPR 2026] Leech Spherical Quantization

Pytorch implementation of stochastically quantized variational autoencoder (SQ-VAE)

Implementation of TiTok, proposed by Bytedance in "An Image is Worth 32 Tokens for Reconstruction and Generation"

A low-bitrate single-codebook 16 / 24 kHz speech codec based on focal modulation

🔥 Official impl. of "DetailFlow: 1D Coarse-to-Fine Autoregressive Image Generation via Next-Detail Prediction"

[official] PyTorch implementation of TimeVQVAE from the paper ["Vector Quantized Time Series Generation with a Bidirectional Prior Model", AISTATS 2023]

Using ideas from product quantization for state-of-the-art neural network compression.

[EMNLP 2024] ESC: Efficient Speech Coding with Cross-Scale Residual Vector Quantized Transformers

Official PyTorch implementation of QuantArt (CVPR2023)

Towards training VQ-VAE models robustly!

Pytorch Implementation of "Neural Discrete Representation Learning"

Dice.com repo to accompany the dice.com 'Vectors in Search' talk by Simon Hughes, from the Activate 2018 search conference, and the 'Searching with Vectors' talk from Haystack 2019 (US). Builds upon my conceptual search and semantic search work from 2015

First open-source implementation of Google TurboQuant (ICLR 2026) -- near-optimal KV cache compression for LLM inference. 5x compression with near-zero quality loss.

Explore how to get a VQ-VAE models efficiently!