Open-source alternatives to K8s Device Plugin

ClearML is a comprehensive MLOps platform designed to manage the end-to-end machine learning lifecycle, from initial experimentation to production deployment. It provides a suite of integrated tools including a pipeline orchestrator for automating workflows, an experiment tracking tool for logging hyperparameters and metrics, and a metadata-driven data versioning system for managing large-scale datasets and model artifacts. The platform is distinguished by its advanced compute management and serving capabilities. It features a GPU compute manager that supports fractional resource slicing and

This project is a Terraform provider that enables the management of Azure cloud resources through an infrastructure as code workflow. It serves as a programmatic interface for creating and configuring services via the Azure Resource Manager API, functioning as a declarative system for provisioning virtual infrastructure. The tool orchestrates the lifecycle of Azure virtual machines, networks, and storage. It replaces manual portal configuration with version-controlled scripts to automate the deployment of cloud services and scale environments across development, staging, and production stages

Devzat is an SSH chat server that provides a real-time communication environment where users connect via a secure shell to participate in chat rooms and terminal-based collaboration. It utilizes the SSH protocol for session management and provides a shared environment for interactive communication through a command-line interface. The project is distinguished by its cross-platform chat bridging, which synchronizes messages between the central server and external platforms such as Slack. It also features an extensible plugin system based on gRPC, allowing for the development of custom bots and

Higgsfield is a distributed machine learning training framework and GPU cluster orchestrator designed for scaling neural networks with billions of parameters. It functions as a large model sharding system and a containerized deployment tool to manage computational workflows across heterogeneous compute resources. The platform provides a centralized interface for experiment management, enabling the monitoring of real-time telemetry, performance metrics, and logs. It ensures reproducible results by using container isolation to standardize dependencies across different computing environments. T

OpenBLAS is a high-performance library for basic linear algebra subprograms that provides optimized matrix and vector operations. It serves as a multi-architecture math backend and numerical computing framework designed to execute complex mathematical calculations and high-speed numerical analysis. The library functions as an optimized CPU math library that detects hardware at runtime to apply the most efficient operation kernels for the specific processor. It supports multiple CPU targets through a combination of optimized assembly and C implementations. The project covers high-performance

gpustack is a GPU cluster management platform and LLM inference orchestrator. It functions as a centralized system for pooling and orchestrating graphics processing units across local servers and cloud environments, serving as a heterogeneous compute manager for diverse hardware and software configurations. The system provides a secure AI model deployment gateway that serves models as scalable services using key-based authentication. It includes a GPU resource scheduler that balances workloads across accelerators and coordinates multiple inference engines to map specific AI models to compatib

Incus is a unified orchestration platform for managing system containers, OCI application containers, and virtual machines through a single control plane. It brings together cluster infrastructure management, secure multi-tenancy, software-defined networking, and pluggable storage backend orchestration into one cohesive system exposed via a full REST API and command-line interface. What distinguishes Incus is its ability to run multiple instance types side by side—full Linux system containers, OCI application containers, and QEMU virtual machines—all managed with consistent tooling. Networkin

Nebullvm is an AI inference accelerator, GPU resource orchestrator, and performance optimization library for large language models. It functions as an optimization layer designed to lower operational costs by aligning model execution with underlying hardware architectures. The system maximizes cluster efficiency through real-time dynamic partitioning and elastic quotas for shared hardware resources. It employs alignment methods and techniques to reduce the hardware and data requirements necessary for tuning large language models. The project covers broad capability areas including AI infrast

Cortex is a Kubernetes-based machine learning infrastructure platform designed for deploying, scaling, and managing models and workloads. It functions as a serverless inference engine and GPU cluster orchestrator, providing the tools necessary to execute real-time, asynchronous, and batch model predictions. The platform utilizes declarative infrastructure-as-code for provisioning model clusters and environments. It optimizes operational costs by elastically scaling CPU and GPU resources through the use of spot instances. The system covers a broad set of operational capabilities, including wo

Runtipi is a home server dashboard and orchestration tool designed for deploying and managing containerized applications. It provides a web-based interface for discovering and installing software from a curated app store, utilizing a Docker Compose orchestrator to handle the deployment of self-hosted services. The system integrates a reverse proxy and SSL manager to route external traffic to internal containers, automating HTTPS certificate renewal and domain assignment. It also features a built-in backup and update manager that uses cron-based scheduling to perform automatic security patchin

Easy-GPU-PV is an administrative toolset for verifying hardware compatibility and automating the deployment of partitioned graphics acceleration across virtualized Windows environments. It functions as a resource orchestrator and manager for provisioning virtual machines with partitioned graphics processing units. The project focuses on Windows GPU partitioning, enabling a single physical graphics card to share hardware acceleration across multiple virtualized systems. It achieves this by automating the configuration of host hardware and software to allow virtualized environments to access gr

kube-bench is a Kubernetes security benchmark scanner and configuration auditor. It verifies if a cluster adheres to the Center for Internet Security standards and other hardening guides to identify security misconfigurations and vulnerabilities. The tool operates as a containerized security scanner, utilizing host namespaces to analyze nodes and control plane components without requiring the installation of binaries directly on the host. It supports multiple Kubernetes distributions, applying environment-specific benchmarks to ensure auditing accuracy for managed services. The project cover

Guides is a collection of standardized architectural guides and best practices for deploying containerized media servers and optimizing download clients. It provides documentation for setting up self-hosted media server environments, focusing on the integration of downloaders and media managers. The project centers on a hardlink storage architecture and a specific directory structure and volume mapping standard. This approach enables atomic moves to prevent data duplication across different media applications. The guides cover a broad range of capability areas, including media automation con

TFLint is a static analysis tool and infrastructure-as-code validator designed to identify errors and enforce best practices within Terraform configuration files. It functions as a quality gate for automated workflows and a language server that provides real-time diagnostics and linting errors directly within supported text editors. The project utilizes a gRPC plugin framework that allows for the integration of custom provider-specific validation rules via independent binaries and a dedicated SDK. This architecture separates core linting logic from cloud-specific validation, supporting the au

Foreman is a lifecycle infrastructure management platform used for automating the provisioning, configuration, and monitoring of physical, virtual, and cloud servers. It serves as a central hub for managing the entire lifespan of a server, from initial deployment and operating system upgrades to decommissioning and auditing. The platform functions as a hybrid cloud manager and bare-metal provisioning tool, providing a unified interface to control virtual machine lifecycles across diverse hypervisors and public cloud providers. It automates hardware discovery and operating system deployment us

Slurm is a cluster workload manager and job scheduler designed for high-performance computing environments. It functions as a distributed compute orchestrator that queues and executes large-scale computational tasks across multiple compute nodes in a cluster. The system acts as a resource arbitrator, distributing hardware nodes and processors among concurrent users to prevent resource conflicts and maximize efficiency. It coordinates the simultaneous launch of multiple processes across different physical servers to execute parallel jobs and scientific workloads. The platform covers broad cap

Accelerate is a framework for high-performance array computing that provides a domain-specific language for expressing complex mathematical and parallel computations. By utilizing a declarative programming interface, it allows users to define high-level array transformations that are automatically translated into optimized machine code for diverse hardware architectures. The system distinguishes itself through a modular architecture that decouples high-level array operations from hardware-specific instructions. It employs just-in-time compilation and kernel fusion to transform programs into e

This project is a technical curriculum and set of educational resources focused on parallel programming, high-performance computing, and systems programming. It provides a structured course covering the implementation of parallel algorithms and multithreading techniques for processing large datasets. The project includes a systems programming guide for modern language features, a framework for lock-free concurrency patterns, and a manual for optimizing CPU and GPU performance through assembly analysis and cache management. The material covers hardware performance tuning, the implementation o

This project is a parallel simulation engine and molecular dynamics simulator designed to model the physical movements of atoms and molecules. It functions as an interatomic potential framework for calculating forces between particles and a materials analysis tool for computing thermodynamic, structural, and transport properties of solids and fluids. The engine is distinguished by its high-performance computing capabilities, utilizing spatial-domain decomposition and message-passing interface communication to distribute workloads across processors. It supports multi-backend GPU acceleration v

Spack is a source-based build system and package manager designed for high-performance computing. It serves as a multi-version software manager and a logic-based dependency solver that handles complex software stacks across various platforms and hardware architectures. The project distinguishes itself by managing multiple compilers and toolchains to target specific hardware. It allows the coexistence of multiple versions and configurations of the same software package on a single system by utilizing prefix-based isolation and unprivileged deployment. The system provides comprehensive capabil

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

kops is a Kubernetes cluster provisioner and lifecycle manager designed to automate the creation, maintenance, and destruction of production-grade clusters on cloud infrastructure. It functions as a declarative infrastructure manager, synchronizing the live state of a cluster with versioned manifests stored in remote object storage to ensure idempotent operations. The project distinguishes itself by offering comprehensive automation for the entire cluster lifecycle, including high-availability control plane deployment, incremental rolling updates, and automated version upgrades. It also serve

ISPC is a vectorizing compiler and SIMD parallel programming language that implements a single program multiple data model. It serves as a toolchain for translating C-based code with parallel extensions into optimized machine code for various CPU and GPU architectures using an LLVM backend. The compiler is designed for cross-platform SIMD toolchain support, generating specialized instruction sets for x86 SSE/AVX, ARM NEON, and Intel GPU from a single source. It features a runtime dispatch mechanism that selects the most efficient hardware-specific implementation for the current system during

Finch is a virtual machine-based container runtime and OCI container management CLI used for local container development. It operates by running container workloads inside a background virtual machine to isolate them from the host operating system. The project serves as an OCI image builder and a multi-container service orchestrator for simulating complex production environments on a workstation. The runtime functions as a cross-platform container engine, utilizing emulation layers to execute container images built for foreign CPU architectures. It distinguishes its image distribution through

Bottles is a Wine compatibility manager and prefix manager that provides a graphical interface for running Windows applications on Linux. It functions as a Windows application sandbox and dependency manager, organizing isolated environments to prevent dependency conflicts and protect the host operating system. The project acts as a Wine runner orchestrator, allowing users to download, install, and switch between different compatibility layers and graphics renderers. It distinguishes itself by using community-driven scripts for automated software installation and dependency management, alongsi

Olares is a comprehensive suite of self-hosted identity, storage, AI, and orchestration services designed for private infrastructure management. It functions as a Kubernetes home server orchestrator, enabling the deployment of containerized applications, AI models, and GPU resources on local hardware to replace third-party cloud services. The platform distinguishes itself through a combination of self-hosted AI infrastructure for running large language models and image generators, alongside a decentralized identity manager that uses cryptographic keys and OIDC for trustless authentication. It