Pricing
Not listed
Official monthly price not confirmed
- API
- Documentation describes HTTP API endpoints for runs, fleets, volumes, secrets, and metrics alongside the CLI.
Finding the possibilities
Dstack
● Identity checkedCode · Vendor site 2026-10-04
dstack is an open orchestration control plane for provisioning and scheduling GPU workloads across clouds, Kubernetes, on-prem clusters, and bare metal. Teams define fleets, dev environments, training tasks, inference services, and volumes in YAML and apply them via CLI or API.
- Category
- Code
- Free access
- Self-hosted dstack server is open source; hosted dstack Sky offers unified GPU cloud access with separate billing
- API access
- Documentation describes HTTP API endpoints for runs, fleets, volumes, secrets, and metrics alongside the CLI.
Is Dstack right for you?
A good fit for
ML engineers and platform teams managing heterogeneous GPU infrastructure who want one interface across vendors and on-prem hardware.
Before you choose
Prices and plan limits were not confirmed from the vendor; check their site before buying.
Heterogeneous AI compute orchestration
dstack markets a single orchestration stack to provision GPUs and run training or inference jobs across any cloud, Kubernetes, or on-prem cluster from shared YAML configs.
What it can do
Features & capabilities
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| Control plane | Docs describe dstack as a unified GPU provisioning and orchestration interface across clouds and on-prem. | Facts sourced | 2026-10-04 |
| Workloads | Configurations include fleets, dev environments, tasks, services, presets, and volumes. | Facts sourced | 2026-10-04 |
| Accelerators | Documentation lists NVIDIA, AMD, TPU, and Tenstorrent support. | Facts sourced | 2026-10-04 |
Understand the total cost
Dstack pricing & plans
Alternatives to Dstack
View all ↗
CerebriumCerebrium is a serverless GPU platform for deploying real-time AI workloads such as voice agents, video models, and LLMs with sub-second cold starts and elastic scaling. Teams run Python apps on managed GPUs across multiple regions and pay for compute by the second rather than reserved capacity.
Explore tool
AmptThe site emphasizes rapid full-stack workflows, built-in CI/CD, preview environments, and pay-as-you-go tiers for solo developers through enterprise self-hosting.
Explore tool
Aikido SecurityAikido Security is a developer-first application security platform covering dependencies, SAST, secrets, cloud misconfigurations, containers, and AI-assisted autofix across the SDLC. Pricing tiers scale from a free Developer workspace to paid Basic, Pro, and Advanced plans for larger repos and infrastructure coverage.
Explore toolThe practical questions
Frequently asked questions
What is Dstack used for?
dstack is an open orchestration control plane for provisioning and scheduling GPU workloads across clouds, Kubernetes, on-prem clusters, and bare metal. Teams define fleets, dev environments, training tasks, inference services, and volumes in YAML and apply them via CLI or API.
Does Dstack have a free plan?
Self-hosted dstack server is open source; hosted dstack Sky offers unified GPU cloud access with separate billing. This record does not confirm an ongoing free plan.
How much does Dstack cost?
A listed monthly price has not been confirmed. See the plan cards for entitlements, billing commitments and seat minimums.
Can I use Dstack through an API?
Documentation describes HTTP API endpoints for runs, fleets, volumes, secrets, and metrics alongside the CLI.. API access and subscription access may have different terms; consult the linked sources.
What should I check before choosing it?
Prices and plan limits were not confirmed from the vendor; check their site before buying.
Price history
No retained pricing changes yet. A current price alone does not establish a historical trend.
How this profile is supported
Identity and profile
Reviewed vendor source
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.