# tHE RAFAY PLATFORM - SERVICES YOU CAN LAUNCH

## Your Infrastructure, Delivered as-a-Service

Rafay provides the foundation for delivering infrastructure “as a service” across private, public, and sovereign environments. From Kubernetes-as-a-Service to Bare Metal-as-a-Service and SLURM-as-a-Service, Rafay lets you define repeatable service blueprints that are governed, multi-tenant, and instantly consumable—no matter where the infrastructure runs.

.webp)

## Launch cloud-like services on any infrastructure

### AI Workbenches

Rapidly experiment with, iterate across, and deploy AI models.

### Landing Zones

Provide all cloud users with self-service access to landing zones using proven templates with guardrails.

### Kubernetes Clusters

Launch fully-compliant Kubernetes clusters in a single click, complete with approval trails.

### SLURM

Deliver SLURM clusters as elastic, multi-tenant HPC services with lifecycle automation and governance.

### Jupyter Notebooks

Offer governed, on-demand JupyterLab environments for data science, AI and ML teams.

### Environments

Enable operations teams and developers to launch guardrails-based environments for immediate use.

### Namespaces

Deliver self-service access to secure, lateral escalation-safe namespaces on-demand using proven templates with guardrails included.

### Serverless Pods

Provide on-demand, customizable compute environments without the overhead of maintaining multiple templates.

### NVIDIA Blueprints

Transform NVIDIA NIM Blueprints into fully operational, self-service AI services.

### Baremetal GPUs

Enable elastic, self-service provisioning of bare metal GPU servers with governance, visibility, and metering built in.

### Models

Deploy, scale, and manage inference endpoints for large language models (LLMs) and other AI workloads.

### NIM-Powered Marketplace

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.

### Inference

Enable providers and enterprises to deploy, scale, and monetize GPU-powered inference endpoints optimized for large language models (LLMs) and generative AI applications.

### Virtual Machines

Deliver GPU- or CPU-based virtual machines as secure, scalable, and consumption-based services.

## Trusted by leading enterprises, neoclouds and service providers

.png)