Mirantis Cloud Platform: AI Infrastructure and MLOps
Mirantis helps enterprises and neocloud providers manage GPU infrastructure and AI workloads. It is designed for organizations building sovereign AI clouds or internal AI factories that require strict multi-tenancy.
At a glance
- Category
- Software Development
- Best for
- Enterprise Companies, Neocloud Service Providers, Government Agencies, Platform Engineers
- Pricing
- Pricing was not clearly available from the provided evidence. Buyers should confirm current pricing on the vendor website.
- Key use cases
- Enterprise AI Factory, Sovereign AI Clouds, AI Infrastructure Monetization, Government AI Infrastructure, Turnkey Inference and Training
- Integrations
- Kubernetes, NVIDIA MIG, NVIDIA vGPU, NVIDIA BlueField
- Official website
- www.mirantis.com
Mirantis provides a suite of infrastructure tools, including the k0rdent AI platform, designed to support the lifecycle of AI workloads. The software handles GPU provisioning and partitioning from bare metal up to the Kubernetes layer, which may help teams deploy training and inference services.
The platform is designed for neoclouds, government agencies, and enterprises that require control over their hardware. It includes features for metering, billing, and hard multi-tenancy to maintain isolation between users or departments.
Buyers should confirm that this platform is intended for those managing their own GPU hardware or building a cloud service, rather than those looking for a plug-and-play AI application.
Mirantis also offers cloud platform operations and Kubernetes management through its Enterprise and open-source k0rdent offerings to support hybrid and edge environments.
Key Features
Supports the allocation of GPU resources using NVIDIA MIG and vGPU to allow multiple teams to share hardware.
Provides isolation across GPU, VM, and Kubernetes layers to support security and compliance requirements.
Includes a marketplace where users can provision GPU, storage, and AI components.
Tracks per-tenant GPU consumption and provides APIs to integrate with billing systems.
Uses declarative templates to support the setup of AI infrastructure from hardware discovery to model deployment.
Supports the deployment and scaling of clusters across data centers, hybrid clouds, and edge locations.
Use Cases
Managing the GPU lifecycle to support moving AI-powered products from experiment to production.
Deploying GPU infrastructure with controls over data residency and regulatory requirements.
Converting underutilized GPU capacity into billable services for external customers or internal departments.
Deploying AI capabilities that align with FIPS and STIG compliance standards for federal or local agencies.
Deploying pre-approved stacks for model training, evaluation, and serving across cloud and edge environments.
Best For
Integrations
Pricing
Pricing was not clearly available from the provided evidence. Buyers should confirm current pricing on the vendor website.
FAQ
It is a multi-tenant AI infrastructure platform that provisions GPUs and core AI services from bare metal to Kubernetes for training and inference.
The platform is designed for enterprise companies, neoclouds, and government agencies building GPU clouds or AI factories.
Yes, it supports GPU partitioning using NVIDIA MIG and vGPU, which may help multiple tenants share hardware.
Source category: Software Development
Source subcategory: MLOps Platform
