simd

ncnn is a high-performance neural network inference framework optimized for the mobile platform

QuestDB is a high performance, open-source, time-series database

A vector index built on TurboQuant, written in Rust with Python bindings

John the Ripper jumbo - advanced offline password cracker, which supports hundreds of hash and cipher types, and runs on many operating systems, CPUs, GPUs, and even some FPGAs

A blazingly fast JSON serializing & deserializing library

A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.

Up to 100x faster strings for C, C++, CUDA, Python, Rust, Swift, JS, & Go, leveraging NEON, AVX2, AVX-512, SVE, GPGPU, & SWAR to accelerate search, hashing, sorting, edit distances, sketches, and memory ops 🦖

The Compute Library is a set of computer vision and machine learning functions optimised for both Arm CPUs and GPUs using SIMD technologies.

Implementations of SIMD instruction sets for systems which don't natively support them.

📽 Highly Optimized 2D / 3D Graphics Math (glm) for C

🚀 efficient approximate nearest neighbor search algorithm collections library written in Rust 🦀 .

High-efficiency floating-point neural network inference operators for mobile, server, and Web

C++ image processing and machine learning library with using of SIMD: SSE, AVX, AVX-512, AMX for x86/x64, NEON, SVE for ARM, HVX for Hexagon

Jlama is a modern LLM inference engine for Java

Llama 3+ inference in pure Java

VectorRAG.Net is a .NET-native high-performance vector database library for semantic search and RAG (Retrieval-Augmented Generation). Core search is based on Random Hyperplane LSH candidate generation with exact rerank by dot/cosine.

The explainable, local-first memory engine for AI agents. One ~9 MB binary fuses vector + graph + columnar under VelesQL; why() returns the evidence path behind every recall. No cloud, no glue code — runs on server, browser, mobile and desktop.

A python library for efficient KNN search within metric spaces using multiple distance functions.

A high-performance Llama3 implementation using Micronaut and GraalVM Native Image

Training-free ordinal & sign quantization for compressed nearest-neighbour retrieval over high-dimensional embeddings. Pure Rust, zero system dependencies.