|
| 1 | +--- |
| 2 | +title: Nebius joins dstack Sky GPU marketplace, with production-ready GPU clusters |
| 3 | +date: 2025-09-18 |
| 4 | +description: "TBA" |
| 5 | +slug: nebius-in-dstack-sky |
| 6 | +image: https://dstack.ai/static-assets/static-assets/images/dstack-sky-nebius.png |
| 7 | +categories: |
| 8 | + - Changelog |
| 9 | +--- |
| 10 | + |
| 11 | +# Nebius joins dstack Sky GPU marketplace, with production-ready GPU clusters |
| 12 | + |
| 13 | +`dstack` is an [open-source :material-arrow-top-right-thin:{ .external }](https://github.com/dstackai/dstack){:target="_blank"} control plane for orchestrating GPU workloads. It can provision cloud VMs, run on top of Kubernetes, or manage on-prem clusters. If you don’t want to self-host, you can use [dstack Sky :material-arrow-top-right-thin:{ .external }](https://sky.dstack.ai){:target="_blank"}, the managed version of `dstack` that also provides access to cloud GPUs via its markfetplace. |
| 14 | + |
| 15 | +With our latest release, we’re excited to announce that [Nebius :material-arrow-top-right-thin:{ .external }](https://nebius.com/){:target="_blank"}, a purpose-built AI cloud for large scale training and inference, has joined the `dstack` Sky marketplace |
| 16 | +to offer on-demand and spot GPUs, including clusters. |
| 17 | + |
| 18 | +<img src="https://dstack.ai/static-assets/static-assets/images/dstack-sky-nebius.png" width="630"/> |
| 19 | +<!-- more --> |
| 20 | + |
| 21 | +Last week we published the [state of cloud GPU](state-of-cloud-gpu-2025.md), a study of the GPU market. As noted there, Nebius is one of the few purpose-built AI clouds delivering performant and resilient GPUs at scale — available on-demand, as spot instances, and as full clusters. |
| 22 | + |
| 23 | +Nebius designs and operates its own GPU servers in energy-efficient data centers, giving full control over quality, performance tuning, and delivery timelines. Every cluster undergoes a three-stage validation — hardware burn-in, reference architecture checks, and long-haul stress tests — ensuring production-ready infrastructure with consistent performance for large-scale AI training. |
| 24 | + |
| 25 | +Since early this year, the open-source `dstack` has supported Nebius, making it easy to manage clusters and orchestrate compute cost-effectively. |
| 26 | + |
| 27 | +## About dstack Sky |
| 28 | + |
| 29 | +With this week's release, Nebius officially joins [dstack Sky :material-arrow-top-right-thin:{ .external }](https://sky.dstack.ai){:target="_blank"}. Nebius can now be used not only with your own account, but also directly via the GPU marketplace. |
| 30 | + |
| 31 | +The marketplace lets you access Nebius GPUs without having a Nebius account. You can pay through `dstack Sky`, and switch to your own Nebius account anytime with just a few clicks. |
| 32 | + |
| 33 | +<img src="https://dstack.ai/static-assets/static-assets/images/dstack-sky-nebius-offers.png" width="750"/> |
| 34 | + |
| 35 | +While the open-source version of `dstack` has supported Nebius clusters from day one, |
| 36 | +Nebius is the first provider to bring on-demand and spot GPU clusters to `dstack` Sky. |
| 37 | + |
| 38 | +With Nebius, `dstack` Sky users can orchestrate NVIDIA GPUs provisioned in hours, with optimized InfiniBand networking to minimize bottlenecks, non-virtualized GPUs for predictable throughput, and industry-leading MTBF/MTTR proven on multi-thousand-GPU clusters. |
| 39 | + |
| 40 | +## Getting started |
| 41 | + |
| 42 | +After you [sign up :material-arrow-top-right-thin:{ .external }](https://sky.dstack.ai){:target="_blank"} with `dstack` Sky, |
| 43 | +you’ll be prompted to create a project and choose between the GPU marketplace or your own cloud account: |
| 44 | + |
| 45 | +<img src="https://dstack.ai/static-assets/static-assets/images/dstack-sky-project-wizard.png" width="750"/> |
| 46 | + |
| 47 | +Once the project is created, install the `dstack` CLI: |
| 48 | + |
| 49 | +=== "uv" |
| 50 | + |
| 51 | + <div class="termy"> |
| 52 | + |
| 53 | + ```shell |
| 54 | + $ uv tool install dstack -U |
| 55 | + ``` |
| 56 | + |
| 57 | + </div> |
| 58 | + |
| 59 | +=== "pip" |
| 60 | + |
| 61 | + <div class="termy"> |
| 62 | + |
| 63 | + ```shell |
| 64 | + $ pip install dstack -U |
| 65 | + ``` |
| 66 | + |
| 67 | + </div> |
| 68 | + |
| 69 | +Now you can define [dev environments](../../docs/concepts/dev-environments.md), |
| 70 | +[tasks](../../docs/concepts/tasks.md), [services](../../docs/concepts/services.md), |
| 71 | +and [fleets](../../docs/concepts/fleets.md), then apply them with `dstack apply`. |
| 72 | + |
| 73 | +`dstack` provisions cloud VMs, sets up environments, orchestrates runs, and handles everything required for development, training, or deployment. |
| 74 | + |
| 75 | +To create a Nebius cluster, for example for distributed training, define the following fleet configuration: |
| 76 | + |
| 77 | +<div editor-title="my-cluster.dstack.yml"> |
| 78 | + |
| 79 | +```yaml |
| 80 | +type: fleet |
| 81 | +name: my-cluster |
| 82 | + |
| 83 | +placement: cluster |
| 84 | +nodes: 2 |
| 85 | + |
| 86 | +backends: [nebius] |
| 87 | + |
| 88 | +resources: |
| 89 | + gpu: H100:8 |
| 90 | +``` |
| 91 | +
|
| 92 | +</div> |
| 93 | +
|
| 94 | +Then, create it via `dstack apply`: |
| 95 | + |
| 96 | +<div class="termy"> |
| 97 | + |
| 98 | +```shell |
| 99 | +$ dstack apply -f my-cluster.dstack.yml |
| 100 | +``` |
| 101 | + |
| 102 | +</div> |
| 103 | + |
| 104 | +Once the fleet is ready, you can run [distributed tasks](../../docs/concepts/tasks.md#distributed-tasks). |
| 105 | +`dstack` automatically configures drivers, networking, and fast GPU-to-GPU interconnect. |
| 106 | + |
| 107 | +To learn more, see the [clusters](../../docs/guides/clusters.md) guide. |
| 108 | + |
| 109 | +With Nebius joining `dstack` Sky, users can now run on-demand and spot GPUs and clusters directly through the marketplace—gaining access to the same production grade infrastrucure Nebius customers use for frontier-scale training, without needing a separate Nebius account. |
| 110 | + |
| 111 | +> If you prefer to go self-hosted, you can always switch to the open-source version of `dstack`, bringing the same functionality. |
| 112 | + |
| 113 | +Our goal is to give teams maximum flexibility while removing the complexity of managing infrastructure. More updates are coming soon. |
| 114 | + |
| 115 | +!!! info "How does `dstack` compare to Kubernetes?" |
| 116 | + `dstack` can run either on top of Kubernetes or directly on cloud VMs. |
| 117 | + In both cases, you don’t need to manage Kubernetes yourself — `dstack` handles container and GPU orchestration, |
| 118 | + providing a simple, multi-cloud interface for development, training, and inference. |
| 119 | + |
| 120 | +!!! info "What's next" |
| 121 | + 1. Sign up with [dstack Sky :material-arrow-top-right-thin:{ .external }](https://sky.dstack.ai){:target="_blank"} |
| 122 | + 2. Check [Quickstart](../../docs/quickstart.md) |
| 123 | + 3. Learn more about [Nebius :material-arrow-top-right-thin:{ .external }](https://nebius.com/){:target="_blank"} |
| 124 | + 4. Explore [dev environments](../../docs/concepts/dev-environments.md), |
| 125 | + [tasks](../../docs/concepts/tasks.md), [services](../../docs/concepts/services.md), |
| 126 | + and [fleets](../../docs/concepts/fleets.md) |
| 127 | + 5. Reaad the the [clusters](../../docs/guides/clusters.md) guide |
0 commit comments