# NVIDIA

## Get results, not just software

Start building today, scale to enterprise tomorrow with guaranteed outcomes every step of the way.

### Community Edition

$00 /forever /month

### Flex Edition

$25,000 / year /month

### Starter Edition

$75,000 / year /month

### Enterprise Edition

$125,000 / year /month

## Core Platform

### Distributed Data Engine
- Locality-aware data operations for RAG

### Inference Mesh
- Scalable inference across environments

### Local Model Repository
- API-accessible model management and deployment

### Data Catalog
- Easy ingestion from files/objects with credential management

## Developer Tools

### Embeddings Middleware
- Model-aware chunking with automatic offset tracking

### Vector Database Access
- Seamless integration with byte-range retrieval

### Cluster Awareness
- Develop on Mac, deploy on Linux clusters

### React UI
- Ready-to-use interface for platform management

## APIs & Integration

### REST APIs
- Comprehensive API suite for custom development

### Jupyter Environment
- Pre-configured with sample notebooks

### Developer Middleware
- Data ingestion, retrieval, and model deployment tools

### Loose Coupling
- Abstracted, fungible architecture for flexibility

## Deployment

### Integrated Stack
- Full Kamiwaza platform with llamacpp/vLLM engine

### Pre-Evaluated Components
- Tested versions of Datahub and dependencies

### Nodes
| Number of cluster nodes |  |  |  |  |
|-------------------------|----|----|----|----|
| 1                       | 3  | 3  | 3  |  |

### CPUs/GPUs
| Number of CPU/GPU Sockets |  |  |  |  |
|---------------------------|----|----|----|----|
| 1                         | +1 | 3  | 3  |  |

### VRAM
| Maximum VRAM per GPU |  |  |  |
|-----------------------|----|----|----|
| -                     | 128GB | 500GB | Unlimited |

## Enterprise Capabilities

### Distributed Processing
- Intelligent workload orchestration across nodes

### Advanced Security Controls
- Enterprise-grade access management

### Production Deployment
- Battle-tested for business-critical applications

### Professional Support
- Priority technical assistance

## Outcome-Based Support

### Guaranteed Outcomes Per Year
- Expert-guided implementation of high-value AI solution

### Dedicated Implementation Support
- Work directly with Kamiwaza engineers

### Solution Design & Optimization
- Custom AI workflow development

### Quarterly Reviews
- Regular assessment of value delivered

### Monthly Reviews
- Regular assessment and optimization

### Dedicated Outcomes Architect
- Assigned strategic implementation partner

### Priority Engineering Access
- Direct line to technical experts

### Strategic Transformation Planning
- Long-term AI roadmap development

NVIDIA and Kamiwaza partner at the forefront of global AI, enabling organizations to operationalize the world’s most advanced AI capabilities.

## Highlights

### Town of Vail to Deploy HPE AI Solution to Make City Smarter, Safer, and More Sustainable

Town of Vail will be the first U.S. municipality to implement HPE’s new agentic AI smart city solution with NVIDIA— enhancing public safety, automating services, and improving everyday life for residents and visitors.

### Kamiwaza Brings Its Enterprise-Grade AI Orchestration Platform to NVIDIA DGX Spark

Kamiwaza v0.8.0 introduces Two-Node Community for DGX Spark, enabling larger-model inference without data center complexity.

### Kamiwaza Releases Its Enterprise-Grade AI Orchestration Platform to NVIDIA DGX Spark

Described as a “data center in a box,” DGX Spark’s full potential is unlocked with Kamiwaza v0.8.0 as it enables users to pool resources across devices seamlessly.

## Featured Solutions

### AI-powered accessibility compliance for government and public sector organizations

A turnkey 508 compliance blueprint from the Unleash AI program.