Awesome GitHub RepositoriesGPU Acceleration Libraries

This project is a high-performance numerical computing library designed for large-scale scientific and machine learning workloads. It functions as an automatic differentiation framework and a just-in-time compilation engine, transforming high-level Python code into optimized machine instructions. By enforcing pure functional programming patterns and immutable array semantics, the library ensures that mathematical functions remain compatible with automated graph transformations and symbolic differentiation. The platform distinguishes itself through its distributed array computing capabilities,

XGBoost is a distributed machine learning library for implementing scalable gradient boosting decision trees used for regression, classification, and ranking. It functions as a predictive model framework and a cross-language toolkit, providing a core implementation with native bindings for Python, R, Java, Scala, and C++. The system is designed as a GPU-accelerated library that utilizes CUDA and NCCL to speed up the training of decision tree ensembles. It operates as a distributed framework capable of scaling training and prediction across multi-node clusters and GPU environments to process m

NVIDIA Docker is a container runtime wrapper that enables the use of host-level graphics processing units within isolated container environments. It functions as a containerized GPU orchestrator, mapping physical hardware resources into virtualized environments to support high-performance computing and machine learning workloads. The project provides a toolkit that facilitates integration between containerized applications and host-level graphics hardware. By utilizing a pre-start hook to intercept container creation, the runtime injects necessary device drivers and libraries into the isolate

This library is a JavaScript framework for general-purpose computing on graphics processing units. It enables the execution of parallel mathematical operations directly within the browser by offloading data-heavy calculations to graphics hardware. The project functions as a web-based math accelerator that converts standard JavaScript functions into shader code for execution on the graphics processor. It provides a unified interface that detects available graphics APIs and manages data transfer between system and graphics memory. To ensure compatibility across diverse environments, the library

Dask es un framework de computación paralela y un programador de tareas distribuido diseñado para escalar flujos de trabajo de ciencia de datos en Python desde máquinas individuales hasta grandes clústeres. Funciona como un gestor de recursos de clúster que orquesta la lógica computacional representando las tareas y sus dependencias como grafos acíclicos dirigidos. Esta arquitectura permite al sistema automatizar la distribución de cargas de trabajo a través del hardware disponible mientras gestiona requisitos de ejecución complejos. El proyecto se distingue por un motor de evaluación perezosa que difiere las operaciones de datos hasta que se solicitan explícitamente, permitiendo la optimización global del grafo y una asignación eficiente de recursos. Incorpora el volcado de datos consciente de la memoria para evitar fallos del sistema al procesar conjuntos de datos que exceden la memoria disponible, y utiliza la fusión de grafos de tareas para combinar secuencias de operaciones en pasos de ejecución únicos, minimizando la sobrecarga de programación y la comunicación entre nodos. La plataforma proporciona una superficie de capacidades integral para el análisis de datos a gran escala, incluyendo soporte para aprendizaje automático distribuido, integración de computación de alto rendimiento y procesamiento de datos en paralelo. Ofrece herramientas extensas para la gestión del ciclo de vida del clúster, perfilado de rendimiento y monitoreo en tiempo real de la ejecución de tareas. Los usuarios pueden desplegar estos entornos en diversas infraestructuras, incluyendo hardware local, proveedores de nube, sistemas en contenedores y clústeres de computación de alto rendimiento.

Cpp-taskflow is a C++ task-parallelism framework and task graph scheduler designed to manage and execute complex dependency graphs of parallel tasks across CPU and GPU hardware. It provides a parallel algorithm library for high-performance implementations of reductions, sorts, pipelines, and iterations. The framework distinguishes itself through its ability to offload heavy computational workloads from a task graph to graphics processors for acceleration. It also includes a task profiling tool and a performance analysis interface for visualizing task execution flow and dependency structures t

CuPy es una biblioteca de computación de matrices CUDA que implementa una interfaz compatible con NumPy para ejecutar operaciones de matrices y computación numérica en GPUs NVIDIA. Sirve como una biblioteca numérica acelerada por GPU y una implementación de SciPy basada en CUDA, descargando cálculos pesados al hardware gráfico para aumentar la velocidad de procesamiento para cargas de trabajo científicas y de ingeniería. La biblioteca permite el intercambio de tensores entre múltiples frameworks, permitiendo que los búferes de datos se compartan entre diferentes frameworks de aprendizaje profundo utilizando diseños de memoria estandarizados para evitar copias de memoria. También admite la integración de kernels de GPU personalizados, permitiendo que los datos de las matrices se conecten a APIs de bajo nivel para un control preciso sobre la ejecución del hardware. En términos generales, el proyecto cubre flujos de trabajo de procesamiento de matrices y computación científica de alto rendimiento. Sus capacidades incluyen la aceleración de cálculos de matrices y la provisión de herramientas para cálculos numéricos a gran escala.

This is a Glide image transformation library for Android that provides a collection of image processing filters and shapes. It functions as a suite of hardware-accelerated tools for image cropping, artistic filtering, and transformation pipeline management. The project distinguishes itself through an image filter pipeline that allows for sequential transformation chaining, enabling multiple visual effects and color modifications to be applied in a single pass. It utilizes GPU acceleration to implement artistic effects such as pixelation, sketching, and blur. The library covers a broad range

jetson-inference is a set of libraries and tools for executing optimized deep learning models on embedded GPU hardware. Its primary purpose is to enable real-time computer vision and AI inference at the edge with low latency and high throughput. The project distinguishes itself through high-performance streaming analytics and the ability to execute concurrent AI pipelines on auto-grade silicon. It provides specialized support for multi-sensor stream processing, utilizing zero-copy data transport to load camera frames directly into GPU memory. The codebase covers a broad surface of capabiliti

PixiEditor is a multi-functional graphics suite that serves as a pixel art editor, a node-based graphics editor, and a vector graphics tool. It functions as a shader-based painting tool and 2D animation software, providing a comprehensive environment for creating raster images and frame-by-frame motion. The project is distinguished by its use of node-based workflows for building complex image transformations, visual effects, and custom digital brush designs. It utilizes a shader-based brush engine and a node graph to define personalized painting tool behaviors and procedural animations. The

GoCV is a computer vision library and Go language binding for OpenCV. It serves as an image processing toolkit and deep learning inference engine, providing programmatic access to a wide range of algorithms for image manipulation, object detection, and video analysis. The project differentiates itself through high-performance native bindings and hardware acceleration. It utilizes a foreign function interface to map Go calls to C++ functions and includes a hardware-agnostic backend dispatch to route neural network tasks to computation engines such as CUDA and OpenVINO. The library covers a br

Feast is an open-source feature store for machine learning that provides a central platform for defining, storing, and serving features across both training and inference workflows. It operates as a declarative system where feature definitions are written as code in Python files, synchronized to a central registry, and made available for low-latency online retrieval or point-in-time correct historical joins for training datasets. The project abstracts storage behind a pluggable architecture, allowing offline and online backends to be swapped without changing retrieval logic, and coordinates ma

Warp is a Python framework that JIT-compiles Python functions into CUDA kernels for GPU-accelerated parallel computation, with built-in automatic differentiation and multi-framework array interoperability. At its core, it provides a GPU kernel compilation system that enables writing and executing custom GPU kernels directly from Python, while supporting automatic gradient computation through those kernels for integration with machine learning pipelines. The framework also includes tile-based cooperative computing, where thread blocks partition into tiles for shared-memory and tensor-core opera

NVIDIA DALI is a GPU-accelerated data loading and preprocessing library designed for deep learning workflows. It constructs high-performance data pipelines that offload decoding, augmentation, and normalization to the GPU, eliminating CPU bottlenecks in training and inference. The library reads data from multiple storage formats and streams it directly into GPU memory, with support for multi-GPU execution to scale throughput across large-scale workloads. DALI distinguishes itself by enabling data pipelines to be built once and executed across multiple deep learning frameworks without code cha

LiteRT-LM es un framework de inferencia de alto rendimiento diseñado para ejecutar modelos de lenguaje de gran tamaño localmente en hardware móvil, de escritorio y IoT. Sirve como un runtime de modelos en el dispositivo que utiliza aceleración de CPU, GPU y NPU para proporcionar un procesamiento de baja latencia. El framework se distingue por su capacidad para procesar entradas de texto, visión y audio a través de un único motor de inferencia multimodal. Cuenta con un servidor HTTP local que emula endpoints de API compatibles con OpenAI y un runtime basado en WebGPU para ejecutar modelos directamente dentro de un navegador web. Para garantizar la fiabilidad de la salida, incluye un generador de texto restringido que impone esquemas JSON o reglas gramaticales en las respuestas del modelo. El proyecto proporciona amplias capacidades para la gestión de conversaciones con estado, decodificación especulativa para aumentar las velocidades de generación de tokens y una interfaz de llamada a herramientas que mapea las solicitudes del modelo a funciones externas. También incluye integración especializada para el ecosistema Apple y un plugin dedicado para ejecutar modelos en Flutter. Los usuarios pueden ejecutar modelos a través de una interfaz de línea de comandos o integrarlos en aplicaciones mediante API nativas.

Flash Linear Attention is a training framework and inference engine for sequence models that use linear attention and state space mechanisms, designed to process long contexts with reduced memory and compute overhead. It provides hardware-optimized token mixing layers and fused CUDA kernels that minimize memory bandwidth and launch overhead across different GPU architectures, and includes a causal inference engine that generates text token-by-token using cached hidden states for efficient autoregressive decoding. The project supports building hybrid sequence models that interleave standard at

TileLang is a Python-embedded domain-specific language compiler that JIT-compiles and autotunes GPU kernels. It uses a tile-based DSL, automatic software pipelining, and parallel autotuning to generate optimized GPU kernels at runtime. It supports tensor core operations with Pythonic syntax, automatic memory management, and thread mapping. The compiler searches over tile sizes, thread counts, and scheduling policies, compiling and benchmarking candidates in parallel to find the fastest kernel. It also caches compiled binaries and tuning results to disk for reuse across sessions. TileLang inc

StableSwarmUI es una interfaz web y orquestador de backend para la generación de imágenes con Stable Diffusion. Funciona como un generador de imágenes GPU distribuido y un pipeline de imágenes de IA modular, proporcionando un controlador centralizado para gestionar las solicitudes de generación de imágenes. El sistema se distingue por la capacidad de dividir las tareas de generación entre múltiples procesadores gráficos para aumentar el rendimiento por lotes. Utiliza una interfaz agnóstica al backend para conectarse a servidores locales, servidores remotos y APIs en la nube, e incluye un diseñador de flujos de trabajo visual basado en grafos para definir operaciones complejas de procesamiento de imágenes. La plataforma incluye un sistema de extensión de plugins dinámico para añadir funciones personalizadas y utilidades automatizadas para el aprovisionamiento de dependencias a nivel de sistema. Combina herramientas de generación modulares e interfaces de edición rápida con la capacidad de enrutar cargas de trabajo a través de hardware distribuido.