simd

A reading path for CPU performance engineering, from one instruction to production inference. Primary sources only, with a runnable benchmark for every section.

Python bindings for the simdjson project.

sigmaker is a zero-dependency IDA Pro 9.0+ cross-platform signature maker plugin with optional SIMD (e.g. AVX2/NEON/SSE2) speedups that works on MacOS/Linux/Windows. It allows configurable wildcard operand patterns, nibbles, and signature generation just by right clicking.

Ai edge toolbox，专门面向边端设备尤其是嵌入式RTOS平台，AI模型部署工具链，包括模型推理引擎和模型压缩工具

Fast Base64 encoding/decoding in Python

Python library for ODE integration via Taylor's method and LLVM

SIMDized check which bytes are in a set

Computing FLOPs with Intel Software Development Emulator (Intel SDE)

Simple example for embedding SSE2 assembly in Cython projects

Ultra-lightweight C++ inference engine for BitMamba-2 (1.58-bit SSM). Runs 1B models on consumer CPUs at 50+ tok/s using <700MB RAM. No heavy dependencies.

Face search engine: 100% recall at 96x less memory than HNSW. Register photos, identify people from camera/video/images. AVX-512 + Rust. No GPU needed.

An O(N) CPU-first linear recurrence inference runtime — hand-written AVX2/AVX-512 kernels, multi-architecture support (SunnyNet / RWKV-7 / Mamba-2 / GLA / DeltaNet / HGRN2)

A tiny approximate K-Nearest Neighbour library in Python based on Fast Product Quantization and IVF

n-body-simulation performance test suite

A fast function (SIMD-accelerated) for finding the minimum and maximum value in a NumPy array

The ApoCLypse Bitcoin Miner

Universal Stochastic Computing Framework for Neuromorphic Hardware — Rust SIMD engine, Python simulation, Verilog RTL, HDC/VSA, SCPN integration

Deterministic symbolic codes for embedding vectors. Byte-identical across CPUs and languages, content-addressed, diffable and verifiable. C core with Python, Rust, Go and TypeScript bindings.

Data-oriented systems programming language for high-throughput, deterministic compute pipelines, enforcing a straight-line SIMD execution model and static memory topology for maximum CPU vectorization.

Build and distribute more platform specific Python extensions