$ loading_
Orchestrate training, inference, and agent workloads across diverse hardware and environments.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "dstack" yet — see the docs or source repo.
Using dstack, create a deployment plan for a PyTorch training job that can switch between AWS and GCP, prefers NVIDIA GPUs, and includes task config, resource selection, launch commands, and retry strategy.
An executable training orchestration plan with cross-cloud resources, run commands, and fault-tolerance settings.
Use dstack to design a large-model inference deployment that supports both Kubernetes and bare metal, can schedule to AMD GPUs or TPUs, and explains scaling and monitoring.
An inference architecture and orchestration recommendation covering multi-environment deployment, scheduling, and operations.
I want to run a set of AI agent tasks with different compute needs. Use dstack to plan the workflow, including task splitting, resource allocation, queued execution, and scheduling between cloud and on-prem clusters.
An agent workload orchestration plan detailing resource needs, execution order, and scheduling mechanisms.
Build, test, and deploy production-ready MCP servers and AI-native apps.
Build intelligent multi-agent dev environments and automate cloud development workflows.
Provides core infrastructure for AI inference, agents, and decision workflows.
Create, edit, run, and publish Python tools as MCP services.
Verify agent deliveries, resolve file conflicts, and return structured refusal reasons.
Create, manage, and run YAML-defined DAG workflows locally for small teams.