NVIDIA Blackwell Ultra AI Factory Platform Paves Way for Age of AI Reasoning

Rhea-AI Summary

NVIDIA has unveiled the next evolution of its Blackwell AI factory platform, called Blackwell Ultra, designed to advance AI reasoning capabilities. The platform includes the GB300 NVL72 rack-scale solution and HGX B300 NVL16 system, delivering 1.5x more AI performance than its predecessor.

The GB300 NVL72 connects 72 Blackwell Ultra GPUs and 36 Arm Neoverse-based NVIDIA Grace CPUs in a rack-scale design. The HGX B300 NVL16 offers 11x faster inference on large language models, 7x more compute, and 4x larger memory compared to the Hopper generation.

Key features include:

• Integration with NVIDIA Spectrum-X Ethernet and Quantum-X800 InfiniBand platforms
• New open-source NVIDIA Dynamo inference framework for enhanced AI services
• Support for agentic AI and physical AI applications

Major tech companies and cloud service providers, including AWS, Google Cloud, and Microsoft Azure, will offer Blackwell Ultra-powered instances starting from the second half of 2025.

Positive

• 1.5x performance improvement over previous generation
• 11x faster inference on large language models
• 50x increase in revenue opportunity for AI factories compared to Hopper
• Significant technological advancement in AI processing capabilities
• Strong partner ecosystem with major cloud providers and manufacturers

Negative

• Extended time to market (availability starting second half 2025)
• High infrastructure requirements for implementation

Insights

AI Infrastructure Expert

NVIDIA's Blackwell Ultra announcement represents a significant leap forward in the company's AI infrastructure portfolio, strategically positioning NVIDIA to capitalize on the emerging AI reasoning market. The new platform delivers 1.5x higher AI performance than its predecessor while expanding Blackwell's revenue opportunity by an impressive 50x compared to Hopper-based AI factories.

The technical specifications are remarkable, with the HGX B300 NVL16 delivering 11x faster inference, 7x more compute, and 4x larger memory than the previous Hopper generation. These performance gains directly address the computational demands of next-generation AI applications including reasoning, agentic AI, and physical AI – all representing expanding market opportunities.

The announcement solidifies NVIDIA's ecosystem dominance through comprehensive partner adoption. All major cloud providers (AWS, Google Cloud, Microsoft Azure, Oracle) and numerous server manufacturers (Dell, HPE, Lenovo, etc.) will offer Blackwell Ultra solutions, creating a robust deployment pipeline scheduled for H2 2025.

The simultaneous introduction of the open-source NVIDIA Dynamo inference framework is particularly notable as it optimizes token revenue generation – a critical metric for AI factory economics – while reducing response times and model serving costs. This software innovation effectively maximizes the ROI for enterprises deploying NVIDIA's hardware.

NVIDIA continues executing its full-stack strategy by providing not just hardware but the comprehensive software ecosystem (NVIDIA AI Enterprise, NIM microservices, CUDA-X) that drives adoption and creates switching costs for customers.

Enterprise AI Strategist

The Blackwell Ultra announcement demonstrates NVIDIA's forward-thinking approach to meeting the next wave of enterprise AI demands. By focusing on reasoning capabilities, NVIDIA is addressing a important barrier to AI implementation – the ability to perform complex problem-solving rather than simple pattern matching.

The test-time scaling approach is particularly innovative, allowing models to leverage massive compute resources during inference to explore multiple solution paths. This directly enables more accurate responses for complex enterprise use cases where reliability is paramount.

For enterprises building AI factories, the 50x increase in revenue opportunity compared to Hopper installations represents a compelling business case for investment. This aligns with the broader enterprise trend toward building proprietary AI infrastructure rather than relying exclusively on API-based services.

The integration with NVIDIA's networking portfolio (Spectrum-X Ethernet and Quantum-X800 InfiniBand) addresses critical enterprise requirements for low latency and consistent performance at scale. The 800 Gb/s throughput per GPU eliminates data movement bottlenecks that currently limit many enterprise AI deployments.

Most importantly, NVIDIA has secured the entire AI supply chain with this announcement – from chips to systems, networking, and software. Enterprise customers gain a unified procurement path through established vendors (Cisco, Dell, HPE) while maintaining access to NVIDIA's innovation. The H2 2025 availability gives enterprises sufficient lead time to budget and plan for implementation while creating a compelling upgrade path from current deployments.

• Top Computer Makers, Cloud Service Providers and GPU Cloud Providers to Boost Training and Test-Time Scaling Inference, From Reasoning to Agentic and Physical AI
• New Open-Source NVIDIA Dynamo Inference Software to Scale Up Reasoning AI Services With Leaps in Throughput, Faster Response Time and Reduced Total Cost of Ownership
• NVIDIA Spectrum-X Enhanced 800G Ethernet Networking for AI Infrastructure Significantly Reduces Latency and Jitter
  

SAN JOSE, Calif., March 18, 2025 (GLOBE NEWSWIRE) -- NVIDIA today announced the next evolution of the NVIDIA Blackwell AI factory platform, NVIDIA Blackwell Ultra — paving the way for the age of AI reasoning.

NVIDIA Blackwell Ultra boosts training and test-time scaling inference — the art of applying more compute during inference to improve accuracy — to enable organizations everywhere to accelerate applications such as AI reasoning, agentic AI and physical AI.

Built on the groundbreaking Blackwell architecture introduced a year ago, Blackwell Ultra includes the NVIDIA GB300 NVL72 rack-scale solution and the NVIDIA HGX™ B300 NVL16 system. The GB300 NVL72 delivers 1.5x more AI performance than the NVIDIA GB200 NVL72, as well as increases Blackwell’s revenue opportunity by 50x for AI factories, compared with those built with NVIDIA Hopper™.

“AI has made a giant leap — reasoning and agentic AI demand orders of magnitude more computing performance,” said Jensen Huang, founder and CEO of NVIDIA. “We designed Blackwell Ultra for this moment — it’s a single versatile platform that can easily and efficiently do pretraining, post-training and reasoning AI inference.”

NVIDIA Blackwell Ultra Enables AI Reasoning
The NVIDIA GB300 NVL72 connects 72 Blackwell Ultra GPUs and 36 Arm
Neoverse-based NVIDIA Grace™ CPUs in a rack-scale design, acting as a single massive GPU built for test-time scaling. With the NVIDIA GB300 NVL72, AI models can access the platform’s increased compute capacity to explore different solutions to problems and break down complex requests into multiple steps, resulting in higher-quality responses.

GB300 NVL72 is also expected to be available on NVIDIA DGX™ Cloud, an end-to-end, fully managed AI platform on leading clouds that optimizes performance with software, services and AI expertise for evolving workloads. NVIDIA DGX SuperPOD™ with DGX GB300 systems uses the GB300 NVL72 rack design to provide customers with a turnkey AI factory.

The NVIDIA HGX B300 NVL16 features 11x faster inference on large language models, 7x more compute and 4x larger memory compared with the Hopper generation to deliver breakthrough performance for the most complex workloads, such as AI reasoning.

In addition, the Blackwell Ultra platform is ideal for applications including:

• Agentic AI, which uses sophisticated reasoning and iterative planning to autonomously solve complex, multistep problems. AI agent systems go beyond instruction-following. They can reason, plan and take actions to achieve specific goals.
• Physical AI, enabling companies to generate synthetic, photorealistic videos in real time for the training of applications such as robots and autonomous vehicles at scale.

NVIDIA Scale-Out Infrastructure for Optimal Performance
Advanced scale-out networking is a critical component of AI infrastructure that can deliver top performance while reducing latency and jitter.

Blackwell Ultra systems seamlessly integrate with the NVIDIA Spectrum-X™ Ethernet and NVIDIA Quantum-X800 InfiniBand platforms, with 800 Gb/s of data throughput available for each GPU in the system, through an NVIDIA ConnectX®-8 SuperNIC. This delivers best-in-class remote direct memory access capabilities to enable AI factories and cloud data centers to handle AI reasoning models without bottlenecks.

NVIDIA BlueField®-3 DPUs, also featured in Blackwell Ultra systems, enable multi-tenant networking, GPU compute elasticity, accelerated data access and real-time cybersecurity threat detection.

Global Technology Leaders Embrace Blackwell Ultra
Blackwell Ultra-based products are expected to be available from partners starting from the second half of 2025.

Cisco, Dell Technologies, Hewlett Packard Enterprise, Lenovo and Supermicro are expected to deliver a wide range of servers based on Blackwell Ultra products, in addition to Aivres, ASRock Rack, ASUS, Eviden, Foxconn, GIGABYTE, Inventec, Pegatron, Quanta Cloud Technology (QCT), Wistron and Wiwynn.

Cloud service providers Amazon Web Services, Google Cloud, Microsoft Azure and Oracle Cloud Infrastructure and GPU cloud providers CoreWeave, Crusoe, Lambda, Nebius, Nscale, Yotta and YTL will be among the first to offer Blackwell Ultra-powered instances.

NVIDIA Software Innovations Reduce AI Bottlenecks
The entire NVIDIA Blackwell product portfolio is supported by the full-stack NVIDIA AI platform. The NVIDIA Dynamo open-source inference framework — also announced today — scales up reasoning AI services, delivering leaps in throughput while reducing response times and model serving costs by providing the most efficient solution for scaling test-time compute.

NVIDIA Dynamo is new AI inference-serving software designed to maximize token revenue generation for AI factories deploying reasoning AI models. It orchestrates and accelerates inference communication across thousands of GPUs, and uses disaggregated serving to separate the processing and generation phases of large language models on different GPUs. This allows each phase to be optimized independently for its specific needs and ensures maximum GPU resource utilization.

Blackwell systems are ideal for running new NVIDIA Llama Nemotron Reason models and the NVIDIA AI-Q Blueprint, supported in the NVIDIA AI Enterprise software platform for production-grade AI. NVIDIA AI Enterprise includes NVIDIA NIM™ microservices, as well as AI frameworks, libraries and tools that enterprises can deploy on NVIDIA-accelerated clouds, data centers and workstations.

The Blackwell platform builds on NVIDIA’s ecosystem of powerful development tools, NVIDIA CUDA-X™ libraries, over 6 million developers and 4,000+ applications scaling performance across thousands of GPUs.

Learn more by watching the NVIDIA GTC keynote and register for sessions from NVIDIA and industry leaders at the show, which runs through March 21.

About NVIDIA
NVIDIA (NASDAQ: NVDA) is the world leader in accelerated computing.

For further information, contact:
Kristin Uchiyama
NVIDIA Corporation
+1-408-313-0448
kuchiyama@nvidia.com

Certain statements in this press release including, but not limited to, statements as to: the benefits, impact, availability, and performance of NVIDIA’s products, services, and technologies; third parties adopting or offering NVIDIA’s products and technologies; Blackwell Ultra being able to easily and efficiently do pretraining, post-training and reasoning AI inference; and advanced networking being a critical component of AI infrastructure that can deliver top performance while reducing latency and jitter are forward-looking statements that are subject to risks and uncertainties that could cause results to be materially different than expectations. Important factors that could cause actual results to differ materially include: global economic conditions; our reliance on third parties to manufacture, assemble, package and test our products; the impact of technological development and competition; development of new products and technologies or enhancements to our existing product and technologies; market acceptance of our products or our partners' products; design, manufacturing or software defects; changes in consumer preferences or demands; changes in industry standards and interfaces; unexpected loss of performance of our products or technologies when integrated into systems; as well as other factors detailed from time to time in the most recent reports NVIDIA files with the Securities and Exchange Commission, or SEC, including, but not limited to, its annual report on Form 10-K and quarterly reports on Form 10-Q. Copies of reports filed with the SEC are posted on the company's website and are available from NVIDIA without charge. These forward-looking statements are not guarantees of future performance and speak only as of the date hereof, and, except as required by law, NVIDIA disclaims any obligation to update these forward-looking statements to reflect future events or circumstances.

Many of the products and features described herein remain in various stages and will be offered on a when-and-if-available basis. The statements above are not intended to be, and should not be interpreted as a commitment, promise, or legal obligation, and the development, release, and timing of any features or functionalities described for our products is subject to change and remains at the sole discretion of NVIDIA. NVIDIA will have no liability for failure to deliver or delay in the delivery of any of the products, features or functions set forth herein

© 2025 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, BlueField, Connect-X, CUDA-X, NVIDIA DGX, NVIDIA DGX SuperPOD, NVIDIA Grace, NVIDIA HGX, NVIDIA Hopper, NVIDIA NIM and NVIDIA Spectrum-X are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and/or other countries. Other company and product names may be trademarks of the respective companies with which they are associated. Features, pricing, availability, and specifications are subject to change without notice.

A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/7bb5b0bf-daad-41dc-8d0f-d1706984d616

FAQ

What performance improvements does NVIDIA's Blackwell Ultra offer compared to previous generations?

Blackwell Ultra delivers 1.5x more AI performance than the GB200 NVL72, with HGX B300 NVL16 offering 11x faster inference, 7x more compute, and 4x larger memory versus Hopper generation.

When will NVIDIA's Blackwell Ultra platform be available to customers?

Blackwell Ultra-based products are expected to be available from partners starting from the second half of 2025.

Which major cloud providers will offer NVIDIA Blackwell Ultra-powered instances?

AWS, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure, CoreWeave, Crusoe, Lambda, Nebius, Nscale, Yotta, and YTL will offer Blackwell Ultra instances.

What are the key technical specifications of NVIDIA's GB300 NVL72 system?

The GB300 NVL72 features 72 Blackwell Ultra GPUs and 36 Arm Neoverse-based NVIDIA Grace CPUs in a rack-scale design, functioning as a single massive GPU.

How does NVIDIA's new Dynamo inference framework enhance AI capabilities?

Dynamo scales up reasoning AI services, improves throughput, reduces response times, and lowers model serving costs by optimizing inference communication across thousands of GPUs.