Awesome GitHub RepositoriesGPU & Performance

This project is a high-level 3D graphics engine designed to render complex, hardware-accelerated environments within web browsers. It provides a comprehensive abstraction layer that manages scene graphs, cameras, and lighting, mapping high-level scene definitions onto low-level graphics APIs. By decoupling these definitions from specific hardware targets, the engine ensures consistent performance across diverse browsers and devices. The framework distinguishes itself through a robust architecture that includes a unified math library for high-frequency spatial calculations and a physically bas

DeepSpeed is a high-performance library designed to scale deep learning model training and inference across massive clusters of GPUs and compute nodes. It provides a comprehensive suite of tools for distributed training, enabling the execution of models that exceed the memory capacity of single devices through advanced parameter partitioning, pipeline-based model parallelism, and memory-efficient state offloading. The framework distinguishes itself through specialized communication-efficient optimizers and hardware-aware acceleration techniques. By utilizing gradient compression, quantization

This project is a high-performance library designed for the similarity search and clustering of dense vectors across massive datasets. It functions as a vector similarity search engine, providing the necessary tools to organize complex numerical data into specialized structures that facilitate rapid retrieval and efficient querying of millions of records. The library distinguishes itself through a variety of advanced indexing and compression techniques, including hierarchical navigable small worlds for logarithmic time complexity and inverted file indexing to partition vector spaces into mana

MediaPipe is a cross-platform machine learning framework designed for deploying vision, audio, and text processing models across mobile, desktop, and web environments. It functions as an on-device inference engine that executes complex models locally on edge hardware, ensuring low latency and privacy without requiring a constant internet connection. The framework utilizes a graph-based pipeline orchestration system where data flows through a directed network of modular calculators to ensure synchronized and deterministic processing. It distinguishes itself through a unified runtime that provi

Frigate is a self-hosted network video recorder that functions as a private, local AI-powered vision engine. It manages video streams by performing real-time object detection, tracking, and classification directly on local hardware, ensuring that security monitoring and activity recording remain independent of cloud services. The system distinguishes itself through a modular, hardware-accelerated video pipeline that offloads intensive decoding and machine learning inference to dedicated GPUs, NPUs, or specialized accelerators like Coral TPUs and Hailo modules. It utilizes state-based object t

Minikube is a command-line tool designed for local Kubernetes development, enabling users to provision and manage full-featured container clusters directly on a workstation. It serves as a local orchestrator that automates the lifecycle of isolated environments, allowing developers to start, stop, pause, and delete clusters to support testing and integration workflows. The project distinguishes itself through its flexible architecture, which supports multiple virtualization drivers and container runtimes to accommodate diverse host environments. It provides deep integration between the host a

Headroom is an AI gateway proxy and token optimizer designed to reduce the cost and latency of large language model interactions. It functions as an intermediary that intercepts traffic between clients and providers to apply context compression, request routing, and format translation. The system differentiates itself through a Model Context Protocol server implementation that delivers compression and retrieval tools to compatible AI hosts. It employs a content-aware compression pipeline and tiered importance scoring to trim redundant data from logs and tool outputs while preserving essential

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

Taichi is a domain-specific programming language embedded in Python designed for high-performance numerical computing and computer graphics. It functions as a parallel compiler that translates high-level mathematical expressions into optimized machine instructions, enabling developers to write compute-intensive algorithms that execute across diverse hardware architectures, including CPUs, GPUs, and specialized accelerators. The project distinguishes itself through a hardware-agnostic execution layer that maps parallel operations to multiple backends such as CUDA, Metal, and Vulkan. By utilizi

This project is a cross-platform graphics and compute framework that provides a unified, hardware-agnostic abstraction layer for rendering and parallel processing. It enables developers to build high-performance applications that execute consistently across diverse operating systems and hardware backends, including Vulkan, Metal, and DirectX. By mapping high-level graphics commands to native APIs, it serves as a portable foundation for both real-time 3D rendering and general-purpose GPU computing. The framework distinguishes itself through a robust architecture that supports both native deskt

bgfx is a cross-platform, graphics rendering abstraction layer designed for high-performance applications. It provides a unified interface that maps high-level rendering commands to native graphics APIs, allowing developers to maintain a single codebase that executes consistently across diverse operating systems and hardware architectures. The library distinguishes itself through a multi-threaded command submission model that decouples rendering logic from the main application thread, effectively minimizing CPU bottlenecks. It utilizes a backend-agnostic command buffer and a deferred resource

This project is a collection of optimized scripts, deployment patterns, and reference implementations designed for scaling and accelerating state-of-the-art AI models. It serves as a multi-domain model zoo and a distributed training framework, providing PyTorch reference implementations for training and deploying models on GPU-accelerated infrastructure. The repository distinguishes itself through an optimization suite focused on NVIDIA GPU hardware, utilizing automatic mixed precision and specialized math modes to increase training speed and throughput. It provides enterprise deployment patt

ZLUDA is a middleware and translation engine designed to enable the execution of unmodified proprietary compute binaries on non-native graphics hardware. It functions as a compatibility layer that bridges vendor-specific compute interfaces with open standards, allowing software originally restricted to a single hardware ecosystem to operate on alternative graphics processing units. The project achieves this through a combination of dynamic library interception and runtime instruction translation. By replacing standard system libraries and mapping proprietary compute calls to open standards, t

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

This project serves as a comprehensive educational framework and reference library for mastering high-performance graphics programming and parallel compute resource management. It provides a collection of practical implementations designed to demonstrate the explicit control required by the Vulkan API, covering the fundamental mechanics of modern graphics pipelines and cross-platform hardware interaction. The repository distinguishes itself by focusing on the low-level architectural requirements of modern GPU development, including manual memory allocation, multi-threaded command recording, a

Numba ist ein Just-in-Time-Compiler, der hochsprachliche Python-Funktionen zur Laufzeit in optimierten Maschinencode übersetzt. Durch die Nutzung der LLVM-Compiler-Infrastruktur bietet es ein Framework zur Beschleunigung numerischer Datenverarbeitung und mathematischer Berechnungen, was Leistungsniveaus ermöglicht, die mit statisch kompilierten Sprachen vergleichbar sind. Das Projekt zeichnet sich durch seine Fähigkeit zur typinferenzbasierten Spezialisierung aus, die Maschinenbefehle generiert, welche auf die während der Ausführung verwendeten Datentypen zugeschnitten sind. Es verwendet eine Lazy-Compilation-Pipeline, die die Übersetzung bis zum Moment des Aufrufs verzögert, wodurch der Start-Overhead minimiert und eine konsistente Leistung über verschiedene Prozessorarchitekturen und Betriebssysteme hinweg beibehalten wird. Über die reine Kompilierung hinaus bietet das Toolkit umfassende Unterstützung für Hardwarebeschleunigung durch die Verteilung iterativer Operationen und Array-Ausdrücke auf mehrere CPU-Kerne und Grafikprozessoren. Es nutzt Vektorisierungs- und Parallelisierungsstrategien, um den Durchsatz für große numerische Datensätze zu maximieren, wodurch Entwickler spezialisierte Hardware direkt aus Standardcode ansprechen können.

Sokol is a C hardware abstraction layer and cross-platform graphics library designed for managing windowing, input, and audio across different operating systems. It functions as a GPU resource manager and multimedia application framework, providing a unified API for rendering 2D and 3D graphics across WebGL, Metal, Direct3D, and OpenGL. The project is distinguished by its single-header implementation, which simplifies integration and portability. It utilizes a stateless render pass definition and a one-update-per-frame model to synchronize CPU data to GPU memory and manage resource lifecycles

This repository is a collection of reference implementations and programming examples for the CUDA Toolkit. It serves as a GPGPU implementation guide and a parallel computing reference, providing code for using graphics hardware to perform general-purpose calculations and high-performance parallel processing. The project provides specific samples for GPU kernel development and resource management. These include demonstrations of multi-GPU communication, peer-to-peer memory access, and system hardware inspection to coordinate distributed GPU resources. The codebase covers a wide range of capa

The PyTorch Tutorials repository is a collection of educational resources that provides step-by-step guidance on building, training, and deploying neural networks using the PyTorch framework. It covers the complete machine learning workflow, from data loading and model definition through optimization loops and model persistence, with dedicated guides for distributed training, model fine-tuning, and deployment. The tutorials offer practical demonstrations of adapting pre-trained models to new tasks through transfer learning, scaling training across multiple GPUs or machines using PyTorch's dis

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