The NVIDIA Blackwell Ultra, the next development of the NVIDIA Blackwell AI factory platform, was unveiled byit, opening the door to the era of AI reasoning
The GB300 NVL72 enhances Blackwell's AI factory revenue potential by 50 times compared to NVIDIA Hopper plants and provides 1.5 times more AI performance than the GB200 NVL72
Neoverse-based NVIDIA Grace CPUs that function as a single, enormous GPU designed for test-time scalability in a rack-scale configuration
The NVIDIA HGX B300 NVL16 outperforms the Hopper generation in AI reasoning with 11x faster inference on massive language models, 7x more computing, and 4x more memory
Physical AI, which enables businesses to produce artificially realistic, real-time videos for the mass training of robots and self-driving cars
NVIDIA ConnectX-8 SuperNIC, Blackwell Ultra NVIDIA systems can interface with Spectrum-X Ethernet and Quantum-X800 InfiniBand platforms, providing 800 Gb/s per GPU
The NVIDIA AI Enterprise software platform for production-grade AI supports the NVIDIA AI-Q Blueprint and new NVIDIA Llama Nemotron Reason models, which work best on Blackwell systems
Blackwell builds on NVIDIA's strong development tools, CUDA-X libraries, more than 4,000 applications that scale performance over thousands of GPUs, and 6 million developers