All Posts

A bold vision — to make distributed computing simple and accessible from anywhere. Open source model, with a developer-first mindset. A vibrant and rapidly growing community. And a stellar team, with a track record of building great technology to sol...

Business logic and model inference have traditionally been handled by different systems. This post describes how Ray breaks down silos by supporting both these workloads seamlessly, allows developers to build and scale microservices as if they were a...

In this post, we’re announcing two new features now stable in RLlib: Support for Attention networks as custom models, and the “trajectory view API”. RLlib is a popular reinforcement learning library that is part of the open-source Ray project.

Double 11 has become the largest online shopping event in the world. To support this level of online activity, Ant Group has implemented a flexible, high-performance, stable, and scalable online resource allocation system based on Ray.

A distributed shuffle is a data-intensive operation that usually calls for a system built specifically for that purpose. In this blog post, we’ll show how a distributed shuffle can be expressed in just a few lines of Python using Ray, a general-purpo...

The pandas library provides easy-to-use data structures like pandas DataFrames as well as tools for data analysis. One issue with pandas is that it can be slow with large amounts of data. It wasn’t designed for analyzing 100 GB or 1 TB datasets. Fort...

Ray is a popular framework for distributed Python that can be paired with PyTorch to rapidly scale machine learning applications.

This blog post highlights two features in the latest Ray 1.2 release: native support for spilling to external storage, and support for libraries from the Python data processing ecosystem, including integrations for PySpark and Dask.

Huggingface Transformers recently added the Retrieval Augmented Generation (RAG) model, a new NLP architecture that leverages external documents (like Wikipedia) to augment its knowledge and achieve state of the art results on knowledge-intensive tas...

This post gives an overview of different ways to speed up your scikit-learn models and discusses some limitations of each approach.