SAN JOSE, CA (GTC), Mar 19, 2025 – NVIDIA has 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 enhances both training and test-time scaling inference by using additional computing power during inference to improve precision. This allows organizations to speed up tasks like AI reasoning, agentic AI, and physical AI.
Built on the Blackwell architecture, 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 and 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 GPU built for test-time scaling. The NVIDIA GB300 NVL72 allows AI models to use its expanded computing capacity to break complex tasks into multiple steps. This helps generate practical solutions and high-quality responses.
GB300 NVL72 is expected to be available on NVIDIA DGX Cloud, an end-to-end 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 give customers a turnkey AI factory.
The NVIDIA HGX B300 NVL16 features 11x faster inference on large language models, 7x more computing, and 4x larger memory compared with the Hopper generation to deliver innovative performance for complex workloads like AI reasoning.
In addition, the Blackwell Ultra platform is ideal for applications including:
- Agentic AI, designed to independently handle complex, multi-step problems by using advanced reasoning and planning. Unlike systems that only follow instructions, these AI models can think, plan, and act to complete specific tasks.
- Physical AI, it allows companies to create synthetic, photorealistic videos in real time to train robots and autonomous vehicles at scale.
NVIDIA Scale-Out Infrastructure for Optimal Performance
Efficient scale-out networking is essential to AI infrastructure, providing robust performance while minimizing delays and disruptions.
Blackwell Ultra systems 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 remote direct memory access capabilities, enabling AI factories and cloud data centers to handle AI reasoning models without disruptions.
NVIDIA BlueField-3 DPUs are also featured in Blackwell Ultra systems, enabling multi-tenant networking, GPU computes elasticity, accelerated data access, and real-time cybersecurity threat detection.
Global Technology Leaders Embrace Blackwell Ultra
Blackwell Ultra-based products will be available from partners starting in the second half 2025.
Cisco, Dell Technologies, Hewlett Packard Enterprise, Lenovo, and Supermicro are set to deliver 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 NVIDIA Blackwell product portfolio is supported by the full-stack NVIDIA AI platform. The NVIDIA Dynamo open-source inference framework – scales up AI reasoning services by increasing processing capacity, reducing response times, and lowering model operation costs with a capable scaling approach for computing during testing.
NVIDIA Dynamo is a 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 fine-tuned separately to meet its requirements, ensuring efficient use of GPU resources.
Blackwell systems are ideal for running 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, 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 development tools, NVIDIA CUDA-X libraries, over 6 million developers, and 4,000+ applications scaling performance across thousands of GPUs.
Source: NVIDIA
About NVIDIA
NVIDIA Corp. is an American tech company headquartered in Santa Clara, CA. Renowned for designing and manufacturing graphics processing units (GPUs), NVIDIA’s innovations have significantly impacted various sectors. The company’s products and services cater to industries such as gaming, where its GPUs enhance visual experiences; artificial intelligence (AI), providing high-performance computing solutions; automotive, contributing to autonomous vehicle technologies; and robotics, offering advanced AI perception and simulation tools. Over its more than three decades in business, NVIDIA has experienced substantial growth. In the fiscal quarter ending January 2025, the company reported record revenue of $39.3 billion and a net income of $22.1 billion. NVIDIA’s headquarters, designed to facilitate a flat organizational structure, emphasizes information flow and harmony between leadership and employees.