New SemiAnalysis InferenceX Data Shows NVIDIA Blackwell Ultra Delivers up to 50x Better Performance and 35x Lower Costs for Agentic AI. Microsoft, CoreWeave and Oracle Cloud Infrastructure are deploying NVIDIA GB300 NVL72 systems at scale for low-latency and long-context use cases.

The NVIDIA Developer Blog provides developers with a  knowledge on GPU computing, AI, and deep learning, offering tutorials, code samples, and real-world applications of NVIDIA technologies. From optimizing GPU-accelerated algorithms to implementing  AI models, developers can learn practical techniques and strategies for harnessing the power of NVIDIA GPUs in their projects. Moreover, the blog highlights advancements in GPU architectures, CUDA programming, and GPU-accelerated libraries, empowering developers to stay at the forefront of GPU computing innovation.

NVIDIA

NVIDIA's Blackwell Ultra platform (GB300 NVL72) delivers up to 50x higher throughput per megawatt and 35x lower cost per token compared to the Hopper platform for agentic AI workloads. The performance gains come from hardware improvements (1.5x higher compute, 2x faster attention processing) combined with continuous software optimizations in TensorRT-LLM, Dynamo, and other frameworks. Major cloud providers including Microsoft, CoreWeave, and Oracle are deploying GB300 NVL72 systems for low-latency and long-context use cases like AI coding assistants. The upcoming Rubin platform promises another 10x improvement in throughput per megawatt for mixture-of-experts inference.

New SemiAnalysis InferenceX Data Shows NVIDIA Blackwell Ultra Delivers up to 50x Better Performance and 35x Lower Costs for Agentic AI

GB300 NVL72 Delivers up to 50x Better Performance for Low-Latency Workloads

GB300 NVL72 Delivers Superior Economics for Long-Context Workloads

NVIDIA Vera Rubin NVL72 to Bring Next-Generation Performance