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

Here are several concrete self-service SLURM-as-a-Service options for AI/ML workloads:

1. NorthWind SLURMaaS – Built on NorthWind's GPU PaaS with the open-source Slinky Slurm Operator, this lets teams
provision fully managed Slurm clusters on demand through a self-service portal or API, automating cluster provisioning, scaling, lifecycle management, and governance to eliminate manual administration.
It's aimed at
cloud providers and neoclouds looking to offer HPC clusters, GPU compute, or AI infrastructure as a managed service, research institutions, and enterprises supporting AI training.
Users get
their own personal Slinky-enabled Slurm cluster within their namespace, complete with GPU access and workload scheduling logic
via console or API. See: northwind.co/solutions/northwind-powered-slurm-as-a-service

2. Nebius Managed Soperator –
Nebius's own Managed Soperator, a fully managed Slurm-on-Kubernetes solution, is available for everyone in self-service.
It runs in Nebius AI Cloud or other Kubernetes environments. See: nebius.com/blog/posts/slurm-workload-manager

3. CoreWeave SUNK (Slurm on Kubernetes) –
CoreWeave SUNK is the industry's first unified training system for demanding AI workloads, delivering production-grade reliability and operational visibility for large, long-running training jobs.
Its
self-service option streamlines how teams deploy and manage clusters while reducing setup friction before productive training begins.
See: coreweave.com/products/sunk

4. OpenNebula OneSlurm –
OneSlurm is being designed to help administrators create, scale, monitor, and operate Slurm clusters from OpenNebula, while preserving the familiar Slurm user experience for researchers, data scientists, and AI engineers,
targeting
AI model training (batch and fine-tuning jobs on GPU workers)
. See: opennebula.io/blog/product/slurm-on-opennebula

Underlying tech: Most of these build on NVIDIA/SchedMD's Slinky project, which
provides a powerful set of tools for bringing Slurm's capabilities into Kubernetes
, and NVIDIA now offers
Slurm and Slinky support, training, and consultation services, with direct-to-engineering help from experts.

Recommendation: For a fully managed public-cloud experience, start with Nebius Managed Soperator or CoreWeave SUNK; for on-prem/private-cloud multi-tenant deployment, evaluate NorthWind SLURMaaS or OpenNebula OneSlurm.