By Anil Murty
Distributed Machine Learning on Akash Network With Ray
In this post we provide an overview of the challenge of trying to parallelize and scale AI and ML workloads in practice, briefly introduce the best open source framework available and in-use by leading machine learning teams (Ray), talk about why Ray is a great solution for running ML workloads on Akash Network, and provide a reference example and template for those looking to run a Ray cluster on Akash Network’s GPU supercloud.
5 Min. Read
By Zach Horn
a/acc: Akash Accelerationism
How strategic progress in 2023 laid the foundation for rapid acceleration in 2024 and beyond.
5 Min. Read
By Akash
Akash Network: Q4 2023 Recap
The demand for compute to power large language models (LLMs) and other applications has surged, resulting in record...
5 Min. Read
By Zach Horn
Building an SDXL Inference App on the Akash Supercloud
In less than two years, AI diffusion models have improved exponentially
5 Min. Read
By Zach Horn
Mainnet 8: GPU Visibility Upgrades
As the proliferation of Large Language Models (LLMs) and AI applications has come to a head, large corporations and agile startups are still struggling to secure access to high-performance GPUs.
5 Min. Read
By Zach Horn
The Supercloud for AI Is Live
Starting today, Akash is launching capabilities for its Supercloud to offer GPUs - The network now supports NVIDIA GPUs, with options for others such as AMD in the future - The network has successfully hosted NVIDIA H100s and A100s, plus a range of consumer-grade GPUs - Access to consumer-grade GPUs is one way the Akash Supercloud stands apart
5 Min. Read
By Zach Horn
Overclock Labs Joins Forces With Cloudmos to Advance the Akash Supercloud
As the world’s demand for compute surges, Akash Network and its community are constantly expanding. The network is always looking for ways to support its most committed contributors, who are invaluable in growing the network’s capabilities.
5 Min. Read
By Zach Horn
Testing the First AI Supercloud
Over 1,300 people from around the world signed up to participate in the Akash GPU Testnet. This three-week public beta test was the first real-world demonstration of GPU compute being distributed with the open-source Akash Supercloud. The Supercloud concept was initially theorized by Cornell University and MIT, before being practically applied in the Akash Network whitepaper.
5 Min. Read
By Zach Horn
Building the Supercloud for AI
The world’s compute needs are increasing exponentially. With the rise of AI and machine learning, chip manufacturers are experiencing unprecedented demand, straining the global supply chain that transforms sand into silicon.
5 Min. Read
By Anil Murty
The Fast Evolving AI Landscape
It’s barely 3 months into 2023 and it is clear that AI is going to dominate the tech space this year, and very likely this decade, by a wide margin (over Web3, Cloud, Mobile or anything else in the past couple decades). If you are in tech
5 Min. Read
By Greg Osuri
Public Cloud Is a Public Utility
The Akash Network was created approximately two years ago to decentralize cloud infrastructure and put it in the hands of the people. It's now run by a network of globally distributed node operators who secure the Akash blockchain and provide cloud-grade server space for tenant applications.
5 Min. Read
By Zach Horn
Mainnet 5: Upgrade instructions, release notes, and more
Akash Network has successfully upgraded to Mainnet 5. The upgrade took place this morning and was completed in ~22 minutes.
5 Min. Read