Find self-service SLURM-as-a-Service options for AI workloads.

Here are the strongest self-service SLURM-as-a-Service options I found for AI workloads:

• NorthWind-Powered SLURM-as-a-Service — a fully managed, multi-tenant offering where tenants can launch SLURM clusters on demand through a portal or API, with per-tenant isolation and governance built in.[1]
• NorthWind + Project Slinky on Kubernetes — enables users to provision their own personal SLURM cluster in minutes via NorthWind’s console or APIs, targeting secure multi-tenant AI/HPC use cases.[2][4]
• Nebius Managed Soperator — Nebius describes this as a fully managed Slurm-on-Kubernetes solution available “in self-service,” aimed at ML and HPC teams running distributed AI workloads.[3][9][16]
• Lambda Managed or Unmanaged Slurm — Slurm job management optimized for AI workloads on Lambda’s 1-Click Clusters and Superclusters; you can choose managed service or self-managed control.[7]
• Google Cloud self-managed Slurm with Cluster Toolkit — not a managed SaaS, but it is a documented self-provisioning path for AI-focused Slurm clusters on B200-backed A4 machines.[14]

If you want the most “true” self-service SaaS experience, the best fits are NorthWind and Nebius Managed Soperator because both explicitly emphasize portal/API-based self-service provisioning.[1][3][16]

Good source URLs to review:

• northwind.co
• northwind.co
• nebius.com
• docs.nebius.com
• lambda.ai