SkyPilot
skypilot.ai
- Free plan
- Linux
- Web
- API
- Self-hosted
What SkyPilot is
SkyPilot is an AI compute platform for managing Kubernetes, Slurm, virtual machines, on-premises systems, and cloud resources through one interface. It covers development, pre-training, post-training and reinforcement learning, batch inference, sandboxes, and inference endpoints. Scheduling includes priority queueing, GPU sharing, quotas, and intelligent placement across workloads. Proactive and reactive fleet health checks can automatically remediate GPU and NCCL-related faults. Listed integrations include AWS, GCP, Azure, Nebius, CoreWeave, Lambda, RunPod, Kubernetes, Slurm, and on-premises infrastructure. SkyPilot can run standalone or connect to a remote API server for multi-user collaboration. The project began as open source at UC Berkeley’s Sky Computing Lab, and an open-source version is available. Enterprise options include SSO, RBAC, workspaces, policy enforcement, cost controls, BYOC, BYOK, private VPCs, and air-gapped deployment. The Trust Center lists SOC 2 Type II. Slurm support remains under active development, and SkyServe deployments on Slurm are not yet supported.
Who it suits
It is aimed at AI teams managing cloud, on-premises, and large GPU workloads across the model lifecycle. Enterprise teams may use its listed identity, policy, deployment, and cost controls.
Good to know, and what to watch
- Manages multiple compute environments through one interface
- Supports priority queues, GPU sharing, and quotas
- Provides GPU and NCCL fault remediation
- Offers broad cloud and cluster integrations
- Includes enterprise deployment and access controls
- Slurm support is under active development
- SkyServe on Slurm is not yet supported
- Enterprise capabilities are listed separately from open source
The verdict
SkyPilot brings workload scheduling, fleet health, and multi-provider compute management into one platform. Teams relying on Slurm should account for the documented deployment limitation while evaluating it.
What the maker publishes
- Compliance
- SkyPilot's Trust Center lists SOC 2 Type II and a 2026 SOC 2 Type II report. source
- Compute integrations
- The maker lists integrations including AWS, GCP, Azure, Nebius, CoreWeave, Lambda, RunPod, Kubernetes, Slurm, and on-premises infrastructure. source
- Cost controls
- The site lists automatic stopping of idle compute, advanced quota management, and cost management and reporting. source
- Cost controls
- The maker lists automatic stopping of idle compute, quota management, and cost reporting. source
- Deployment model
- The documentation says SkyPilot can run standalone or connect to a remote API server for multi-user collaboration. source
- Enterprise security
- The site displays SOC 2 Type II and describes BYOC, BYOK, private VPC, and air-gapped deployment options. source
- GPU health
- It provides proactive and reactive fleet health checks with auto-remediation for GPU and NCCL-related faults. source
- Integrations
- The site lists Kubernetes, Slurm, AWS, GCP, Azure, Nebius, CoreWeave, Lambda, RunPod, and other cloud and compute providers. source
- Open source
- SkyPilot began as an open-source project at UC Berkeley's Sky Computing Lab, and the site links to an open-source version. source
- Open source
- SkyPilot began as an open-source project at UC Berkeley's Sky Computing Lab. source
- Product
- SkyPilot unifies Kubernetes, Slurm, VMs, and on-premises compute to run AI workloads across the AI lifecycle. source
- Purpose
- SkyPilot is an AI compute platform for managing Kubernetes, Slurm, VMs, on-premises systems, and cloud resources through one interface. source
- Scheduling
- SkyPilot supports priority queueing, GPU sharing, quotas, and intelligent scheduling across workloads. source
- Scheduling
- SkyPilot offers workload scheduling, GPU monitoring, quotas, priority queueing, GPU sharing, and GPU health checks with auto-remediation. source
- Security deployment
- The enterprise offering lists bring-your-own-cloud and bring-your-own-key, private VPCs, and air-gapped deployment. source
- Slurm limitation
- The documentation says Slurm support is under active development and SkyServe deployments on Slurm are not yet supported. source
- Support
- The maker invites users to connect with its team and community on Slack and offers a demo request for the platform. source
- Support
- The maker invites users to connect with the SkyPilot team and community through Slack. source
- Target users
- The maker positions SkyPilot Platform for frontier AI teams managing large GPU fleets and AI workloads. source
- Team controls
- Enterprise capabilities listed include SSO onboarding, a unified compute dashboard, policy enforcement, RBAC, and workspaces. source
- Team controls
- The maker lists SSO onboarding, role-based access control, workspaces, policy enforcement, and a unified compute dashboard. source
- Usage limit
- The maker describes SkyPilot Platform as optimized for large GPU fleets and frontier workloads, with performance to scale to thousands or tens of thousands of GPUs. source
- Workloads
- The site lists development, pre-training, post-training and reinforcement learning, batch inference, sandboxes, and inference endpoints as use cases. source
- Workloads
- The platform supports AI development, pre-training, post-training and reinforcement learning, batch inference, sandboxes, and inference endpoints. source
Checked against the maker's pages on 5 Oct 2026.