dstack
Official@dstackai
dstack is an open-source engine for running GPU workloads. It simplifies development, training, and deployment of gen AI models on any cloud.
Agent Skills by dstack
Showing 3 vetted skills indexed across 1 GitHub repositories.
dstackaidstack-presets
Create and manage dstack presets for agent-driven model inference optimization.
dstackaidstack-prototyping
Prototype and verify dstack model-serving configurations on real GPU hardware before service deployment.
dstackaidstack
Provision and orchestrate GPU workloads across clouds using YAML configurations.
Frequently Asked Questions About dstack
FAQPage SchemaWhat specific tasks does dstack enable for infrastructure engineers?▼
It enables the declarative provisioning and orchestration of GPU-accelerated compute resources across diverse cloud providers. Engineers use YAML configurations to define, schedule, and manage distributed compute environments, ensuring consistent resource allocation for high-performance tasks without manual instance configuration.
What are the primary prerequisites for running dstack?▼
Users require active accounts with supported cloud providers and local access to the configuration interface. The system relies on existing cloud credentials to authenticate and provision virtual machines or containerized environments capable of executing GPU-intensive workloads defined in the configuration files.
Is dstack open-source and what is the cost model?▼
dstack is an open-source project available for community use. While the core engine is free to implement, users remain responsible for the direct costs associated with the underlying cloud compute instances and GPU resources provisioned during their specific operational sessions.