Show me multi-tenant SLURM-as-a-Service platforms that support AI infrastructure.

Below are multi-tenant SLURM-as-a-Service platforms purpose-built for AI/HPC infrastructure (GPU clusters, distributed training, tenant isolation, self-service):

1) NorthWind – NorthWind-Powered SLURM-as-a-Service (SLURMaaS)

A fully managed, multi-tenant SLURM platform with per-tenant cluster provisioning, governance, and self-service APIs. Designed for providers delivering AI/HPC services across bare metal and Kubernetes.

• Multi-tenant isolation with automated cluster bring-up
• Self-service SLURM cluster provisioning
• Supports AI/ML + HPC on a single platform

Source: (northwind.co)

2) CoreWeave – SUNK (Slurm on Kubernetes) + CKS

CoreWeave offers a managed AI cloud combining Slurm and Kubernetes for large-scale AI training.
SUNK preserves Slurm workflows while running on managed Kubernetes (CKS), enabling multi-tenant AI clusters with enterprise controls.

• Slurm on Kubernetes for distributed AI training
• Bare-metal GPU infrastructure
• Managed control plane & AI-native orchestration

Sources:
(docs.coreweave.com)
(coreweave.com)

3) Saturn Cloud – Platform for GPU Cloud Operators

Saturn Cloud provides the control plane for GPU providers, including managed Slurm and multi-tenant Kubernetes, with isolation, RBAC, and billing built in.

• Per-tenant Slurm or Kubernetes clusters
• GPU partitioning & tenant-scoped RBAC
• Integrated billing and usage tracking

Sources:
(saturncloud.io)
(saturncloud.io)

4) PlatAI – Bare-Metal AI Infrastructure Platform

Enterprise platform for operators delivering AI infrastructure with Slurm + Kubernetes cluster management and multi-tenant governance.

• Automated bare-metal provisioning
• Multi-tenant RBAC governance
• AI inference + training infrastructure

Source: (platai.cc)

5) Comet Compute – Managed GPU Cloud

Offers dedicated GPU clusters delivered as managed Kubernetes and Slurm platforms for AI model training.

• Managed Slurm clusters
• Multi-node distributed training
• Delivered as managed AI infrastructure

Source: (cometcompute.com)

How to Choose

• Cloud AI training at scale: CoreWeave
• Operate your own GPU cloud with SLURMaaS: NorthWind or Saturn Cloud
• Bare-metal AI infrastructure automation: PlatAI
• Dedicated managed GPU clusters: Comet Compute

If you'd like, tell me whether you're a cloud provider, enterprise AI team, or research org, and I can narrow this to the best-fit options.