Open-source alternatives to Microsoft Pai

OpenLLM is a framework for deploying, managing, and scaling open-source large language models

BytePS is a distributed deep neural network training framework and communication library designed to scale model training across multiple GPUs and compute nodes. It functions as a GPU cluster orchestrator and RDMA network optimizer, providing the necessary primitives to synchronize gradients and data across a server cluster. The project distinguishes itself through high-performance network optimizations, utilizing remote direct memory access and page-aligned memory to reduce latency. It employs topology-aware communication tuning and CPU core affinity management to maximize hardware throughpu

Open framework for confidential AI

Horovod is a distributed deep learning framework and gradient synchronizer designed to scale model training across multiple GPUs and compute nodes. It functions as a distributed training orchestrator and an elastic training engine, utilizing an MPI collective communication library to synchronize weights and gradients across TensorFlow, PyTorch, Keras, and MXNet models. The system distinguishes itself through dynamic elastic scaling, which allows it to adjust the number of active workers at runtime and recover from node failures. It optimizes communication efficiency using tensor fusion batchi

🚪✊Knock Knock: Get notified when your training ends with only two additional lines of code

Sacred is an experiment management tool and reproducibility framework designed to organize multiple runs of a process with different configurations. It functions as a machine learning experiment tracker and hyperparameter configuration manager, logging hyperparameters, metrics, and metadata to a database to ensure that experimental executions remain trackable. The project focuses on scientific result reproducibility by automatically managing random seeds and tracking system dependencies. It allows for the execution of experiment variants through command-line parameter overrides and dynamic pa

open-source MLOps platform

DVC is a data versioning tool and pipeline orchestrator designed to track large datasets and machine learning models. It functions as a system for managing large data artifacts by storing lightweight metadata in version control while keeping the actual binaries in a separate cache. The project serves as an experiment tracker and remote storage synchronizer, enabling the execution and comparison of machine learning iterations based on hyperparameters and performance metrics. It provides a bridge for pushing and pulling these large data artifacts between local environments and cloud or on-premi

KServe is a Kubernetes-native platform for deploying and serving machine learning models as scalable inference services. It supports both generative AI models, including large language models, and traditional predictive models from frameworks such as TensorFlow, PyTorch, Scikit-Learn, XGBoost, and ONNX. The platform manages the full lifecycle of model deployments, including revision tracking, canary rollouts, A/B testing, and automatic rollbacks, and provides serverless scale-to-zero capabilities for cost-efficient resource management. KServe distinguishes itself through a standardized infere

Kubernetes-native Job Queueing

Hopsworks - Data-Intensive AI platform with a Feature Store

NLP DNN Toolkit - Building Your NLP DNN Models Like Playing Lego

Debugging, monitoring and visualization for Python Machine Learning and Data Science

ModelFox makes it easy to train, deploy, and monitor machine learning models.

Metaflow is a Python machine learning framework and MLOps workflow orchestrator designed to manage the lifecycle of data pipelines from local prototyping to production. It serves as a distributed compute manager and an experiment tracking system, enabling the creation of reproducible pipelines that transition between development and high-availability production environments. The framework distinguishes itself through an integrated checkpointing system that automatically persists intermediate data artifacts to remote storage, allowing failed runs to be resumed from the last successful step. It

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

DIGITS is a GPU deep learning training platform and model manager used to train, fine-tune, and manage neural network models on NVIDIA hardware. It functions as a REST-controlled machine learning pipeline that integrates with S3 cloud storage for dataset ingestion and organization. The platform supports image classification workflows, allowing users to train various model architectures and export trained image classifiers for use in external environments. It includes capabilities for model fine-tuning to adapt pretrained weights to specific tasks. The system provides a REST-based API interfa

Optuna is a Python-based hyperparameter optimization framework designed to automate the search for optimal machine learning model configurations. It functions as a Bayesian optimization library that systematically tests parameter combinations to maximize or minimize objective functions, streamlining the model development process through iterative evaluation. The project distinguishes itself through a define-by-run dynamic construction model, which allows users to build complex, conditional search spaces using standard programming logic. Its architecture is highly modular, featuring a pluggabl

VisualDL is a deep learning visualization toolkit and experiment tracking dashboard. It provides a web-based interface for monitoring training metrics, analyzing high-dimensional data, and rendering model architectures through static and dynamic graphs. The toolkit serves as a performance profiler to identify execution bottlenecks and optimize resource usage. It also functions as a data analyzer that uses projection algorithms to identify relationships between points in complex datasets. Capabilities include tracking training metrics via scalars and histograms, comparing multiple experiments

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

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

Seldon Core is a Kubernetes-based machine learning model server and MLOps inference framework. It functions as a multi-model serving engine and pipeline orchestrator, packaging models as scalable microservices that are exposed via standardized REST and gRPC APIs. The project distinguishes itself through graph-based inference pipelines that chain models and data transformers into sequential workflows. It optimizes hardware utilization via multi-model shared serving and dynamic memory overcommit strategies, while supporting production experimentation through weighted traffic routing, A/B testin

Machine Learning Platform and Recommendation Engine built on Kubernetes

an MLOps/LLMOps platform