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

Here are the leading multi-tenant SLURM-as-a-Service platforms built for AI infrastructure:

1. NorthWind-Powered SLURM-as-a-Service (SLURMaaS) – The clearest purpose-built offering.
NorthWind-powered SLURM as a Service (SLURMaaS) enables organizations to deliver fully managed, multi-tenant SLURM environments for high-performance computing (HPC) workloads.
It runs
SLURM by integrating it with Kubernetes through Project Slinky, allowing providers to expose SLURM job scheduling as a cloud-like, on-demand service.
Key capabilities include
self-service access where tenants can launch SLURM clusters instantly through a portal or API, Kubernetes integration via Project Slinky for containerized HPC workloads, and fully automated lifecycle management (provisioning, scaling, patching, teardown).
A companion architecture combines Slinky with NorthWind's GPU PaaS to give
enterprises and cloud providers a transformative combination that enables secure, multi-tenant, self-service access to Slurm-based HPC environments on shared Kubernetes clusters.

→ northwind.co
→ northwind.co

2. Slinky (SchedMD's SLURM-on-Kubernetes project) – The underlying open-source technology many vendors build on. Multiple GPU cloud providers use it to deliver managed multi-tenant SLURM:

• Crusoe has built this into a production offering, noting:

"Slurm is the standard for large-scale GPU training, but operating it as a managed cloud service is a different problem entirely," covering how Crusoe built Slurm on Managed Kubernetes using Slinky v1.0.

→ crusoe.ai

• CoreWeave offers Slurm on Kubernetes (SUNK) for GPU-heavy AI batch infrastructure, described as fitting best where

containerized environments, elastic capacity, and multiteam self-service matter.

→ coreweave.com

3. OpenNebula (Slurm Appliances / AI Factory Blueprints) – A more open-infrastructure route. OpenNebula's blueprints will
make it easier for AI Factories and HPC centers to offer Slurm-based training environments as part of a broader AI service portfolio, alongside GPU-as-a-Service, Kubernetes-based AI workloads.

→ opennebula.io

4. NVIDIA Slurm – The upstream engine now co-developed by SchedMD engineers at NVIDIA, offering
support on‑prem, cloud, and hybrid deployments,
which underpins most of the above SaaS offerings.
→ nvidia.com

Recommendation: If you want a turnkey, sellable multi-tenant SLURM service today, NorthWind's SLURMaaS is the most explicitly packaged product. If you're a GPU cloud/neocloud building your own service, Slinky-based deployments (as done by Crusoe and CoreWeave) are the reference architecture to evaluate.