Operationalizing AI Fabrics with Aviz ONES, NVIDIA Spectrum-X, and Rafay
Discover the new AI operations model available to enterprises that enables self-service consumption and cloud-native orchestration for developers.
The NIM-Powered Model Marketplace as a Service, delivered by Rafay, enables telecommunications providers and sovereign cloud operators to transform NVIDIA GPU infrastructure into fully operational AI marketplaces.
NVIDIA NIM (NVIDIA Inference Microservices) are containerized, performance-optimized microservices that run AI models on NVIDIA GPUs and expose industry-standard APIs. The Rafay Platform is an infrastructure orchestration and automation platform that enables self-service consumption of compute and AI services.
Built on NVIDIA NIM and orchestrated by Rafay, this solution allows telcos to launch branded AI marketplaces where enterprises can select, deploy, and consume AI services instantly using self-service, with full governance, sovereignty, and monetization controls.
Static model catalogs lack operational readiness, custom integrations delay launches, and enterprises need sovereign, compliant services.
Rafay and NVIDIA NIM solve this by enabling telcos to stand up governed, production-ready marketplaces in weeks, connecting compute, models, and blueprints in a unified platform.
.webp)
Rafay provides the orchestration and automation layer needed to operationalize AI service delivery for telcos, enterprises, and sovereign cloud providers.
White-labeled portals allow enterprises to browse, deploy, and consume AI services instantly.
Built-in usage tracking, compliance, and sovereignty ensure every deployment meets enterprise and regulatory standards.
OpenAI-compatible APIs enable applications to access AI models through standardized endpoints.
Telcos can stand up fully branded AI marketplaces in weeks, drastically reducing time to production.
NVIDIA NIM (NVIDIA Inference Microservices) provides packaged inference capabilities for AI models. Rafay supports NVIDIA NIM alongside other AI technologies. The platform allows organizations to run NIM-powered models on GPU systems and make them available as governed services through the Rafay Platform.
Rafay enables organizations to provision and manage endpoints for LLMs and other AI workloads running on GPUs. Providers can use NIM-powered models as part of a catalog of AI services. Self-service access, tenant-level governance, usage tracking, and infrastructure orchestration can be managed through the Rafay Platform.
Providers can offer NVIDIA NIM-powered Models-as-a-Service, NVIDIA Blueprints, and Compute-as-a-Service with Kubernetes clusters, virtual machines, and GPU slices—all through a catalog that extends beyond GPU capacity alone.
Rafay exposes NIM-powered models as governed API endpoints. Customers can deploy available services through a self-service experience and use the resulting endpoint to send requests to the model and receive a response, while providers maintain tenant-level governance and usage visibility.
Yes. The Rafay Platform can be consumed as SaaS or deployed in self-hosted environments, and Rafay supports NIM and AI service delivery across private, sovereign, and air-gapped environments. This allows providers to operate AI services while maintaining control over their systems and data.
Rafay's NIM documentation lists a shared host Kubernetes cluster with GPUs as a prerequisite for its NIM Microservices template. Administrators configure the required service profile and model settings, including an NGC API key, before making the service available to end users through the Rafay catalog. Detailed configuration requirements are available in the Rafay product documentation.
No. Rafay provides the orchestration, automation, governance, and service-delivery layer for AI infrastructure. NVIDIA and other providers supply the underlying GPU hardware and AI technologies. Rafay enables providers to turn those resources into governed, self-service services that customers can consume.
Yes. Rafay supports NIM-powered models as API endpoints, allowing applications and users to interact programmatically with deployed inference services. Rafay also provides APIs and other programmatic interfaces that providers can use to integrate platform capabilities into their existing portals and service-delivery environments.
Talk with Rafay experts to assess your infrastructure, explore your use cases, and see how teams like yours operationalize AI/ML and cloud-native initiatives with self-service and governance built in.
Learn how to Streamline Kubernetes Ops in Hybrid Clouds with AWS & Rafay
