dstack is an open-source control plane for AI infrastructure orchestration. It enables provisioning and managing GPUs across clouds, Kubernetes, and on-prem clusters for training, inference, and development.
dstack is an open-source control plane designed for agentic orchestration of AI infrastructure. It allows engineers to provision compute and run training, inference, and sandboxes across NVIDIA, AMD, TPU, and Tenstorrent GPUs on various platforms including cloud providers, Kubernetes, and bare-metal clusters. dstack simplifies the management of AI workloads by offering a unified control plane for compute orchestration, eliminating the need for complex Kubernetes or Slurm configurations. It supports running development environments, batch jobs, and high-performance model inference with features like auto-scaling, resource allocation, and GPU health metrics. dstack aims to reduce GPU costs through efficient reuse, right-sizing, and support for different capacity types.
Best used for
Ideal for developers and MLOps engineers who need to provision compute resources, run training and inference jobs, and manage development environments across diverse GPU infrastructure. Especially valuable for streamlining AI workflows and optimizing GPU utilization without the complexity of Kubernetes or Slurm.
Not publicly disclosed. Check dstack.ai for current pricing.
FAQs
How does dstack compare to Kubernetes for AI workloads?
dstack offers a lightweight, ML-native interface specifically for AI, simplifying GPU provisioning and cluster management. While Kubernetes is a general-purpose orchestrator, dstack focuses on dev environments, tasks, and services for ML, and can even run on top of existing Kubernetes clusters if needed.
Can dstack manage GPUs on bare-metal servers?
Yes, dstack can connect to your existing bare-metal servers or pre-provisioned VMs using SSH fleets. You provide SSH credentials and host addresses, and dstack creates an SSH fleet to manage your compute resources for development, tasks, and services.
What types of GPUs does dstack support?
dstack supports a wide range of accelerators including NVIDIA, AMD, TPU, and Tenstorrent GPUs. It provides out-of-the-box compatibility for provisioning and orchestrating these diverse hardware types across various cloud and on-premise environments.
Shypd.aiThe largest open index of AI tools
