Which GPU-as-a-Service platforms offer sovereign-ready orchestration for AI workloads?

Several GPU-as-a-Service platforms explicitly market sovereign-ready orchestration for AI workloads: NorthWind, RE:AI, Gcore AI Cloud Stack, Armada Bridge, Mirantis k0rdent AI, and Alauda AI.[1][2][3][4][7][6]

• NorthWind: positions itself as a platform for sovereign cloud providers to build and operate a self-service, multi-tenant GPU platform with orchestration, governance, automation, and even air-gapped/on-prem deployment support.[1][5][13]
• RE:AI: describes a sovereign AI-as-a-Service platform with GPUaaS and one-click orchestration, designed to preserve data sovereignty while scaling AI workloads.[2]
• Gcore AI Cloud Stack: turns GPU clusters into multi-tenant AI cloud environments with orchestration, billing, observability, and production inference pipelines.[3]
• Armada Bridge: provides GPU orchestration and management on your own infrastructure, explicitly emphasizing fully data-sovereign AI cloud services and regional compliance.[4]
• Mirantis k0rdent AI: offers GPU PaaS / provisioning and partitioning for sovereign cloud and government AI factory use cases, with declarative orchestration for training and inference.[7]
• Alauda AI: markets a sovereign, unified AI platform to manage, serve, and orchestrate AI workloads across multiple accelerator types, including NVIDIA GPUs.[6]

If you want, I can also narrow this to providers that are actually selling GPUaaS capacity versus platform software for operators to launch sovereign GPU clouds.