simdjson is a high-performance C++ JSON parsing library that uses SIMD instructions (AVX2, AVX-512, NEON, RVV, etc.) to parse and validate JSON at multi-gigabyte-per-second speeds. It provides two primary APIs: a DOM API (`simdjson::dom`) that builds a complete in-memory tree, and an On-Demand API (`simdjson::ondemand`) that performs lazy, forward-only parsing for streaming and large-document use cases. The library also includes a builder API for JSON serialization, compile-time JSON parsing via C++26 reflection, JSON Pointer/Path support, NDJSON/JSON-sequence streaming, and a single-header amalgamated distribution. The codebase is organized around a two-stage pipeline: stage 1 performs SIMD-accelerated structural indexing and UTF-8 validation, and stage 2 builds the DOM tape or drives the on-demand iterator. Runtime CPU dispatch selects the best implementation for the host architecture, with fallbacks for non-SIMD platforms. The repository is a mature, production-grade library used by major projects (Node.js, ClickHouse, etc.) and is backed by an extensive test suite, fuzzing harnesses, and cross-architecture CI. It is primarily an application/library rather than a data-research repository, though it contains benchmark and research-oriented components (e.g., performance modeling, static reflection experiments).
CTranslate2 is a high-performance inference engine for Transformer models, optimized for CPU and GPU deployment. It supports a wide range of architectures including encoder-decoder (BART, T5, NLLB, M2M-100, Whisper), decoder-only (LLaMA, GPT-2, Mistral, Gemma, Qwen), and encoder-only (BERT, DistilBERT, XLM-RoBERTa) models. The system provides a C++ library with Python bindings, a command-line tool (ct2-translator), and converters from popular frameworks like HuggingFace Transformers, Fairseq, Marian, OpenNMT-py/tf, and OPUS-MT. Key features include quantized inference (INT8/INT16/FP16/BF16/AWQ), flash attention, multi-GPU tensor parallelism, streaming generation, and token-level penalties. The architecture follows a modular design with device-agnostic operations, layer abstractions, and a model registry that allows plugging in new architectures without modifying the core engine. The codebase is actively maintained with frequent releases, extensive test coverage (C++ and Python), and CI pipelines for multiple platforms and backends.
Corrfunc is a high-performance scientific computing library for cosmology, specifically for computing two-point correlation functions and the void probability function from galaxy catalogs. It provides both a C library/command-line interface and Python bindings through C extensions. The code implements pair counting for real-space (DD, xi, wp), redshift-space (DDrppi, DDsmu), angular (DDtheta), and mock/survey geometries, with support for periodic simulation boxes and line-of-sight survey data. The architecture is a hybrid C core with Python wrappers: OpenMP parallelized C kernels, runtime dispatch across SSE4.2/AVX/AVX2/AVX512 intrinsics, grid-based cell decomposition, and a custom fast-food binary format for fast I/O. The project also includes a substantial test suite with reference outputs, CI workflows, and benchmark scripts used in a peer-reviewed performance paper. Despite being packaged as an application/library, the repository contains research artifacts: benchmark .npz files, paper scripts, and astrophysical mock-catalog test data. These are support materials for validating and advertising performance, not the primary deliverable. The dominant character is clearly a reusable scientific library, not a data-analysis pipeline. The codebase is mature and well-engineered for its niche, but shows some drift: Python 2 compatibility remnants, a complex dual Makefile/setup.py build, heavy code duplication between theory and mock modules, and platform-specific SIMD code that requires careful maintenance. It remains actively maintained and is the de facto standard CPU tool for fast correlation functions in cosmology.
BitNet is a high-performance inference framework for 1-bit large language models, specifically the BitNet b1.58 ternary architecture. It provides CPU (x86 with AVX2, ARM with NEON) and GPU (NVIDIA CUDA) backends with custom quantization formats (I2_S, TL1, TL2) that pack ternary weights and use efficient look-up tables and integer matrix multiplication (int8×int2) to achieve significant speedups over full-precision models (1.37–6.17× on CPU). The system is built on a fork of llama.cpp, adding custom GEMM kernels and a Python utility suite for model conversion, embedding quantization, benchmark, and deployment. It includes a CLI interactive mode and an HTTP server for production serving. The repository contains extensive performance benchmarks and configuration tools, making it a production-ready solution for deploying 1-bit LLMs on commodity hardware.
This repository provides a header-only library, NEON_2_SSE.h, that enables porting of ARM NEON intrinsics code to x86 platforms by mapping ARM NEON intrinsics to equivalent x86 SIMD intrinsics (SSE, AVX2, etc.). The library is designed to be used as a drop-in replacement for arm_neon.h when compiling for x86, allowing developers to write SIMD code using NEON intrinsics that can be compiled on both ARM and x86 without modification. The project includes a CMake-based build system for installation and packaging, and the header is self-contained, requiring only the target platform's compiler intrinsics headers. The codebase appears to be stale, with the latest copyright dates in the provided files (LICENSE, CMakeLists.txt) indicating updates up to 2020. The provided last commit date (2026-04-28) is invalid (future date), suggesting a data error, but the file copyrights imply no recent activity. The library focuses on compatibility up to AVX2 and does not show signs of technical drift, as it relies on stable compiler intrinsics interfaces. The primary value lies in the NEON_2_SSE.h header, which implements a non-trivial mapping between SIMD architectures.