Fast inference engine for Transformer models

A quantum puzzle and adventure into Native Language decolonization; features an introduction to the master quantum plane and the truthful history of indigenous peoples. Not G-rated.

Repo for AMX + FAST

A tool that performs layer-wise quantization of LLM models to optimize the quality-to-size tradeoff beyond uniform quantization methods. Built on top of 'llama.cpp'.

A lightweight library for the RaBitQ algorithm and its applications in vector search.

Support fixed posit quantised training, inference and fine tuning of neural networks (pytorch based) using the highly optimised fp multiplication on GPU

Quantized BLAS

Faiss-based library for efficient similarity search

A header-only library for serializing and quantizing bits

[SIGMOD 2024] RaBitQ: Quantizing High-Dimensional Vectors with a Theoretical Error Bound for Approximate Nearest Neighbor Search

[SIGMOD 2025] Practical and Asymptotically Optimal Quantization of High-Dimensional Vectors in Euclidean Space for Approximate Nearest Neighbor Search

A simple BMP image viewer, converter and editor. App is primarily focused on implementation of own code for working with BMP images

A resource-conscious neural network implementation for MCUs

An attempt to process through the MoE quantization issues found when converting from safetensors to GGUF

INT4/INT5/INT8 and FP16 inference on CPU for RWKV language model

LinearCosine: Adding beats multiplying for lower-precision efficient cosine similarity

My Undergraduate Thesis: EGEMV on Intel SoCs

TinyChatEngine: On-Device LLM Inference Library

C++ Image Manipulation Program. Edit your photos by swapping colours, quantise colours or flip the image and more.

최적의 Scalar uniform/non-uniform 양자화기 및 역양자화기 설계