
Open source control plane that provisions cheap GPUs across clouds
Do you work on dstack?
This listing is unclaimed. Claim it for free to edit the details, add screenshots and reply to reviews.
Machine learning teams use dstack to run training and inference jobs on whichever GPUs are cheapest. It is an open source orchestrator that works across several clouds, Kubernetes clusters and bare-metal servers. Teams describe GPU requirements declaratively, and dstack provisions resources from providers such as AWS, GCP, Azure and Lambda or from on-premises hardware. That hides cloud-specific APIs and cuts the cost of LLM development. A free tier is listed.
From people who have used it. One review per person, checked by a human.
No reviews yet. Be the first to say something useful.