Download
Cube Studio is an open-source, cloud-native end-to-end machine learning and AI platform designed to support the full lifecycle of AI development — from data preparation and interactive notebook coding to distributed training, model tuning, and deployment in production-ready environments. It provides a unified interface where teams can manage data sources, track datasets, and build pipelines using drag-and-drop workflow orchestration, making it accessible for both engineers and data scientists working at scale. The platform supports distributed training across multiple machines and GPUs, integrates tools for automated hyperparameter search and logging, and can serve models via inference services that include virtualized GPU support for efficient utilization.
Features
- End-to-end cloud-native machine learning and AI platform
- Notebook-based development and interactive coding environment
- Drag-and-drop pipeline orchestration and workflow automation
- Multi-node distributed training with hyperparameter search
- Inference services with GPU virtualization and edge deployment
- Support for many ML/DL frameworks and ecosystem tools
Project Samples
