Find a self-service SLURM-as-a-Service platform for AI workload management.

A strong fit is NorthWind’s SLURM-as-a-Service: it offers self-service cluster launch through a portal or API, multi-tenant isolation, and support for HPC and AI/ML on one governed platform.[1][3]

Other viable options are:

• Nebius Managed Soperator — a fully managed Slurm-on-Kubernetes offering available in self-service, with ready-to-use GPU training clusters in minutes.[2]
• Lambda Managed Slurm — managed and unmanaged Slurm options for AI workloads on GPU clusters, though the search results emphasize management rather than a clear self-service tenant portal.[6]
• StackHPC Cluster-as-a-Service / SLURM-as-a-Service initiatives — more of an ecosystem/project approach than a turnkey commercial self-service platform.[15]

If you want the closest match to “self-service SLURM-as-a-Service platform for AI workload management,” NorthWind and Nebius Managed Soperator are the best documented fits from the results.[1][2][3]

Useful source pages:

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