News

User Story

Engineering

Culture

Younes Abouelnagah

image2

Roblox Guest Blog: Fast and Efficient Online Model Serving

Kevin Su

Kai-Hsun Chen

Learn how we combine the power of Anyscale and Union to easily construct a batch inference pipeline.

image

Building a RAG Batch Inference Pipeline with Anyscale and Union

Marwan Sarieddine

Screenshot 2024-05-06 at 3.11.06 PM

Kamil Kaczmarek

image4

Fine-tuning Llama-3, Mistral and Mixtral with Anyscale

An end-to-end guide to transitioning from a pandas-based implementation to a scalable Ray Data solution. Get started with Ray Data today.

image3

From Pandas to Ray Data: Scaling Embedding Generation Pipelines

Scaling Embedding Generation Pipelines From Pandas to Ray Data

Franklin Wang

Sumanth Hegde

Kourosh Hakhamaneshi

A case study of Direct Preference Optimization (DPO) with synthetic data on Anyscale

preference-tuning

How To Do Direct Preference Optimization on Anyscale

Direct Preference Optimization with Synthetic Data on Anyscale

Build and enhance your GenAI and LLM skills at Ray Training Day. Level up your knowledge with an intensive, in-person day of training led by Ray experts.

hero banner

Ray Training Day 2024: September 30th in San Francisco

Don’t Miss: Hands-On Ray Training at Ray Summit 2024

Robert Nishihara

Anyscale appoints industry veteran Keerti Melkote chief executive officer following 6x year over year growth of Ray capturing the generative AI boom.

Anyscale - Welcome Keerti

Anyscale Appoints Keerti Melkote Chief Executive Officer

Welcome Keerti

​​Reinventing Multi-Modal Search with Anyscale and MongoDB

Dominic Catalano

Alexey Kudinkin

Learn how to use Anyscale Job Queues to execute multiple Ray jobs on the same cluster. Job Queues make it easy to improve utilization and reduce costs.

The lifecycle of a job queue cluster on Anyscale.

image1

Introducing Anyscale Job Queues

Improve Utilization and Simplify Cluster Management with Anyscale Job Queues

Matthew Deng

Justin Yu

Ensure continuous ML job execution with Anyscale’s new elastic training. Automatically adapt to changes in resources, even with hardware failures or spot node preemptions.

Introducing Elastic Distributed Training: Reduce Costs up to 60% by Running on Spot

Introducing Elastic Distributed Training on Anyscale

Amjad Almahairi

Reduce LLM application costs by as much as 2X. Get started with our out-of-the-box LLM router template. Easily train and deploy on Anyscale.

LLM Router Template – Reduce LLM Costs by up to 2X

Now Available: The LLM Router Template

Alan Guo

Gene Su

With the Unified Log Viewer access and search logs for any cluster in Anyscale over the last 30 days. Learn how to get started with the Unified Log Viewer today.

Screenshot 2024-07-18 at 9.13.53 AM

Introducing Anyscale’s Unified Log Viewer

This section is used to order the "Types" and "Tags" that show up for filters on the Blog Index

Types

Products / Libraries

Blog

Anyscale is the leading AI application platform. With Anyscale, developers can build, run and scale AI applications instantly.

Anyscale

Featured Posts and News

Building a RAG Batch Inference Pipeline with Anyscale and Union

Fine-tuning Llama-3, Mistral and Mixtral with Anyscale

Scaling Embedding Generation Pipelines From Pandas to Ray Data

No posts found.