AI Storage Ecosystem for the Data Center

# AI Storage

Accelerated storage infrastructure for AI workloads, built with partners.

[Read Announcement](https://nvidianews.nvidia.com/news/nvidia-vera-bluefield-4-stx-brings-agentic-ai-storage-processing-with-in-silicon-security)

Overview

## Foundation for the New Frontier of AI Storage

NVIDIA is empowering partners to build the first AI-native storage solutions, transforming passive data into active fuel for [AI factories](https://www.nvidia.com/en-us/solutions/ai-factories.md). While traditional solutions were designed for general-purpose computing, agentic AI requires a new class of storage. By integrating accelerated computing and [networking](https://www.nvidia.com/en-us/networking.md) into the storage fabric and optimizing the software stack, we help partners deliver hyperscale-grade efficiency across inference and training. This foundation allows organizations to turn their massive datasets into immediately accessible intelligence.

### NVIDIA BlueField-4 STX Storage Platform Launches With Broad Industry Adoption

NVIDIA STX reference architecture, co-designed with leading storage partners, enables enterprises and cloud and AI providers to easily deploy accelerated storage infrastructure capable of the long-context reasoning required for agentic AI.

[Read Press Release](https://nvidianews.nvidia.com/news/nvidia-launches-bluefield-4-stx-storage-architecture-with-broad-industry-adoption)

### NVIDIA Vera BlueField-4 STX Brings Agentic AI Storage Processing With In-Silicon Security

Embed agent, memory, and storage protection in silicon with NVIDIA® Vera BlueField®-4 STX, enforcing real-time, policy-driven security at AI agent speed.

[Read Press Release](https://nvidianews.nvidia.com/news/nvidia-vera-bluefield-4-stx-brings-agentic-ai-storage-processing-with-in-silicon-security)

Architectures

## Explore Our AI Storage Architectures

### NVIDIA STX

* Modular reference architecture for AI storage workloads
* Co-designed with leading storage partners
* Built on NVIDIA accelerated compute, networking, and AI software

[Foundation for AI-Native Storage Platforms](https://www.nvidia.com/en-us/data-center/ai-storage/stx.md)

### NVIDIA CMX™ Context Memory Storage

* AI-native storage tier extending GPU context memory for long-context inference
* Enables high-speed sharing of key-value (KV) cache across rack-scale nodes
* Improves throughput and power efficiency over traditional storage

[Context Memory Storage](https://www.nvidia.com/en-us/data-center/ai-storage/cmx.md)

### NVIDIA AI Data Platform

* Customizable reference design that integrates NVIDIA accelerated computing into enterprise storage
* Centralizes intelligent data handling and delivers AI-ready data
* Reduces latency, enhances data security, and maximizes performance

[AI Data Platform for Enterprise](https://www.nvidia.com/en-us/data-center/ai-data-platform.md)

### NVIDIA-Certified Storage

* Enterprise storage from leading NVIDIA partners
* Certified to deliver performance, security, and scale required for production AI workloads
* Enables faster model training, lower latency inference, and better infrastructure utilization

[Accelerated Storage for the AI Era](https://www.nvidia.com/en-us/data-center/products/certified-storage.md)

Ecosystem

## AI Storage Partner Ecosystem

Resources

## Learn More About AI Storage

[See All](https://resources.nvidia.com/l/en-us-ai-storage)

## Next Steps

### Ready to Get Started?

Explore the future of enterprise storage, powered by NVIDIA AI and accelerated computing.

[Find a Partner](https://marketplace.nvidia.com/en-us/enterprise/partners/?location=US&page=1&limit=15)

### Get in Touch

Speak with an NVIDIA product specialist to explore collaboration on our enterprise storage solutions and the implementation of partner storage solutions.

[Contact Us](https://www.nvidia.com/en-us/contact.md)

### Stay Up to Date on NVIDIA AI News

Get the latest AI news, technologies, breakthroughs, and more sent straight to your inbox.

[Stay Informed](https://www.nvidia.com/en-us/ai-data-science/generative-ai/news.md)