Open-source alternatives to Volcano

FedML is a distributed machine learning training library, federated learning framework, and GPU workload orchestrator. It provides the core system components necessary to execute large-scale model training and fine-tuning across multi-cloud, on-premise, and decentralized GPU clusters, while offering a dedicated engine for scalable model serving and an MLOps pipeline manager for end-to-end lifecycle management. The platform distinguishes itself by enabling privacy-preserving federated learning across decentralized edge devices and organizational silos, keeping raw data on local hardware. It al

Bull is a Node.js library for managing distributed jobs and message queues using Redis as the primary data store. It functions as a distributed task worker, job scheduler, and priority queue manager designed to handle asynchronous workloads across multiple processes. The project distinguishes itself by providing a persistent communication channel that decouples servers through the exchange of serializable data objects. It ensures distributed system reliability by detecting stalled tasks and recovering from process crashes to ensure every queued job is completed. The system covers a broad ran

Polyaxon is a Kubernetes-native machine learning orchestration platform and MLOps pipeline orchestrator. It serves as a control plane for managing distributed deep learning workloads, automated machine learning pipelines, and experiment tracking. The platform distinguishes itself through specialized services for distributed training management, including MPI-based coordination for PyTorch and TensorFlow. It provides an automated hyperparameter optimization service utilizing Bayesian, random, and grid search algorithms, alongside managed interactive AI workspaces for launching Jupyter notebook

Hatchet is an open-source durable workflow engine and task orchestration platform. It provides a framework for building and executing fault-tolerant, multi-step pipelines as directed acyclic graphs (DAGs), with automatic retries, scheduling, and real-time observability. The system is built around durable task checkpointing, which persists execution state after each step so work can resume from the last checkpoint after a worker crash or restart, and it supports event-driven task resumption that pauses a task until a matching external event arrives. The platform distinguishes itself through it

Maestro is a distributed job scheduler and containerized data pipeline tool designed to coordinate complex sequences of tasks. It functions as a Kubernetes workflow orchestrator and MLOps automation platform, utilizing directed acyclic graphs to manage task dependencies and execution order across computing clusters. The system distinguishes itself through the use of isolated container environments for each workflow step, ensuring consistent runtime dependencies. It incorporates an asynchronous event bus to coordinate state transitions and provides lifecycle hook integration that dispatches sy

Enterprise job scheduling middleware with distributed computing ability.

Oban is a distributed background job processing system and task scheduler that uses PostgreSQL for transactional job storage and reliable execution across multiple nodes. It serves as a PostgreSQL-backed background worker and job queue, coordinating task execution and concurrency through a relational database to ensure delivery guarantees. The system differentiates itself through a distributed workflow orchestrator capable of managing multi-step processing pipelines, dependent job sequencing, and shared context. It provides advanced orchestration tools including job batching, chunked processi

s1 is a reasoning training framework and GPU cluster orchestrator designed to build and refine large language models. It provides a system for executing supervised fine-tuning on distributed hardware, utilizing gradient checkpointing and hardware optimization to improve model reasoning. The project features a synthetic data generator and dataset builder that produce high-quality training sets. This workflow collects questions, generates model reasoning traces, and applies automated grading loops to filter for correct answers. The framework includes an evaluation suite to compute accuracy and

RLinf is a distributed reinforcement learning orchestrator and embodied AI training framework. It provides the infrastructure to train vision-language-action models and robotic policies using a combination of reinforcement learning and supervised fine-tuning. The system is designed for scaling workloads across GPU clusters, managing the placement of actors, rollout workers, and environment components. It features a specialized robotics data collection pipeline for gathering teleoperated demonstrations and simulation trajectories into standardized replay buffers, alongside a hardware interface

This project is a comprehensive engineering framework and technical reference for managing, scaling, and optimizing distributed machine learning infrastructure. It provides a suite of methodologies and diagnostic tools designed to support large-scale model training and inference on high-performance computing clusters. The project distinguishes itself through a specialized diagnostic toolkit and infrastructure optimization suite that addresses the complexities of multi-node environments. It enables precise control over cluster resources, including hardware maintenance, network topology configu

This project provides a comprehensive technical guide and framework for engineering large-scale machine learning systems. It covers the full lifecycle of model development, focusing on the infrastructure and computational principles required to build, train, and serve generative AI models across distributed GPU clusters. The repository distinguishes itself by offering deep-dive tutorials and implementation strategies for complex system challenges. It emphasizes high-performance architectural primitives, such as collective communication orchestration, distributed tensor sharding, and static gr

This project is a distributed training infrastructure designed for aligning large language models through reinforcement learning. It functions as an end-to-end engine for complex alignment tasks, including proximal policy optimization, direct preference optimization, and iterative self-play. By providing a unified framework for multi-turn interactions and tool-use scenarios, it enables the development of models capable of reasoning and external environment engagement. The framework distinguishes itself through a decoupled architecture that separates model training from sample generation. This

Search-R1 is a distributed training system and reinforcement learning framework designed to create search-augmented language models. It provides an architecture for scaling model workloads across head and worker nodes while optimizing how models interleave internal reasoning with external tool calls. The system focuses on refining model behavior through custom reward signals and reinforcement learning to improve tool-use formatting and information retrieval. It implements an interleaved reasoning-search loop that allows models to alternate between internal thought generation and external data

This project is a comprehensive educational resource and curriculum focused on the design and implementation of the full machine learning software and hardware stack. It serves as a technical reference for architecting machine learning systems, spanning from low-level programming interfaces to large-scale deployment infrastructure. The project provides instructional guidance on several specialized domains, including the development of AI compilers through intermediate representations and graph optimizations. It covers the architectural patterns required for distributed training across GPU clu

This project is a standardized machine learning experiment boilerplate and project template that combines PyTorch Lightning with the Hydra configuration framework. It provides a structured codebase for organizing deep learning workflows, specifically designed to integrate hierarchical configuration management with distributed training. The template features a specialized workflow for hyperparameter optimization and batch experiment execution, allowing for automated parameter sweeps without modifying source code. It employs a hierarchical system for managing settings via YAML files and command

Baselines is a comprehensive suite of frameworks for reinforcement learning algorithm implementation, imitation learning, and training orchestration. It provides a library of standardized learning algorithms used to benchmark and replicate research results, alongside a deep learning policy framework for constructing neural network architectures such as multi-layer perceptrons, convolutional networks, and long short-term memory networks. The project includes a specialized imitation learning toolkit that enables agents to mimic expert behavior through behavior cloning and generative adversarial

BullMQ is a Redis-backed message queue library and background processor designed for distributed task queueing. It functions as a distributed queue manager and task scheduler, utilizing Redis to manage asynchronous job processing and persistence. The system distinguishes itself through its role as a job workflow orchestrator, enabling the definition of complex parent-child job dependencies and hierarchies for multi-step workflows. It provides sandboxed process execution to isolate heavy workloads and prevent event loop blocking, alongside distributed rate limiting to protect downstream servic

Faktory is an open-source work server that queues, dispatches, and manages background jobs across multiple programming languages. It stores job payloads as JSON hashes in a Redis-backed queue and provides language-specific client and worker libraries that enable any language to push jobs to the server or fetch and execute them. The server includes a batch workflow orchestrator that groups jobs into batches with completion tracking for coordinating multi-step asynchronous workflows. It features a configurable job uniqueness filter that prevents duplicate enqueues within a time window, an expon

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

AI-System is an educational resource and toolkit designed for learning the hardware and software foundations of deep learning systems. It provides a curriculum and practical exercises for building AI infrastructure, ranging from low-level CUDA kernel development to high-level system management. The project includes a toolkit for developing tensor operations and optimizing GPU performance through direct hardware programming. It also features a framework for distributed training, focusing on resource scheduling and communication protocols to manage large-scale models across multiple computing n

NNI is an AutoML toolkit designed to automate machine learning lifecycles. It functions as a hyperparameter optimization framework, a neural architecture search tool, and a model compression suite. The project provides a distributed training orchestrator to manage machine learning workloads across local machines, remote servers, and cloud platforms. It enables the discovery of efficient model structures through reinforcement learning and one-shot optimization methods, while utilizing Bayesian and evolutionary algorithms to automate hyperparameter tuning. Additional capabilities include tools

Torchtune is a PyTorch-native library for fine-tuning, aligning, and quantizing large language models. It provides a configurable training pipeline orchestrated through YAML recipes, with CLI overrides and component swapping, distributed training via FSDP2, memory optimizations, and parameter-efficient fine-tuning methods like LoRA, DoRA, and QLoRA. The library distinguishes itself through its YAML-driven configuration system that defines all training parameters and instantiates components from config files, with full CLI override capability for any field or component at launch time. It suppo

Torchtune is a PyTorch-native library for fine-tuning, aligning, and quantizing large language models. It provides a config-driven system for instantiating components, orchestrating distributed training, and managing parameter-efficient fine-tuning with quantization support, all through YAML-based configurations and command-line overrides. The library distinguishes itself through its comprehensive post-training workflow orchestration, combining supervised fine-tuning, preference optimization (DPO, PPO, GRPO), knowledge distillation, and quantization-aware training in a single configurable pip

Composer is a PyTorch distributed training framework designed for scaling large-scale models across multi-node GPU clusters. It functions as a large language model trainer, a distributed model optimizer, and a training lifecycle manager. The project differentiates itself as a deep learning regularization library, providing specialized optimization techniques such as Sharpness Aware Minimization, MixUp, and CutMix to improve model generalization. It further distinguishes its training flow through the use of sequence length warmup, progressive layer freezing, and sharded-state checkpointing for

Kue is a Redis-backed job queue library for Node.js that provides a complete system for defining, scheduling, and processing background work. It stores job metadata and state in Redis lists and sorted sets, enabling persistent, in-memory operations with configurable concurrency control and priority-sorted processing. The library includes a RESTful HTTP API for managing jobs and a web-based monitoring dashboard for inspecting job status, progress, and logs. The system distinguishes itself through its event-driven worker model, where workers listen for job events via Redis pub/sub and process j

PyTorch Lightning is a high-level deep learning framework for PyTorch that automates training loops and removes repetitive engineering boilerplate. It functions as a structured pipeline for managing machine learning experiments, providing a distributed training orchestrator and tools for mixed-precision training. The framework decouples scientific model architecture from the engineering required for infrastructure and scaling. This separation allows the same model code to execute across CPUs, GPUs, or TPUs through a hardware-agnostic execution engine and a centralized trainer that manages the