AcademySkillsDevelopment

dstack

Engineering and data teams that need to run AI training and inference on different GPU infrastructure from one place.

  • Recommendedour rating
  • 2,276gitHub stars
  • MPL-2.0licence
  • 2 days agolast update
git clone https://github.com/dstackai/dstack.git
README.md

Loading the file...

What it does

dstack is a control plane for provisioning and orchestrating GPU compute. It works with GPU clouds, Kubernetes, VMs and on-prem clusters, and covers development, training and inference. It supports NVIDIA, AMD, Google TPU and Tenstorrent accelerators.

Use cases

  1. 01Run model training across several GPU clouds
  2. 02Serve inference on a Kubernetes cluster
  3. 03Manage GPU capacity on bare-metal servers

Questions about
dstack.

What is dstack used for?

Engineering and data teams that need to run AI training and inference on different GPU infrastructure from one place. dstack is a control plane for provisioning and orchestrating GPU compute. It works with GPU clouds, Kubernetes, VMs and on-prem clusters, and covers development, training and inference. It supports NVIDIA, AMD, Google TPU and Tenstorrent accelerators.

How do I install dstack?

Run this in your terminal: git clone https://github.com/dstackai/dstack.git

Is dstack open source?

Yes. The code is on GitHub (dstackai/dstack) under the MPL-2.0 licence.

Is dstack safe to use?

It passed our automatic scan for credential theft, hidden instructions and risky install commands. Third party open source software. Sabemos AI does not maintain it. Check the code and permissions before you connect it to company data.