B200s now generally available.
Cirrascale Cloud Services has added the Nvidia HGX B200 GPUs to its cloud offering.
The B200s are now generally available via Cirrascale's "AI Innovation Cloud."
The HGX B200 platform features Nvidia NVLink, Nvidia NVSwitch, and Nvidia BlueField‑3 DPU technologies. For multi-trillion parameter models, it delivers up to 5× faster real‑time inference throughput and up to 12× lower cost‑per‑inference than the Hopper generation.
“Integrating the Nvidia HGX B200 into our AI Innovation Cloud lets innovators accelerate trillion‑parameter models and generative‑AI deployments with unprecedented scale and efficiency,” said Mike LaPan, VP of Marketing at Cirrascale Cloud Services. “Our customers can now move from experimentation to production faster, while controlling cost and performance.”
“Nvidia HGX B200 brings record‑breaking performance and scalability to cloud infrastructures,” said Dave Salvator, director of accelerated computing products at Nvidia. “Through Cirrascale, organizations can harness Blackwell’s capabilities to train and deploy next‑generation models faster than ever.”
Cirrascale announced plans for an inference-as-a-service platform that will use Nvidia B200s and the Nvidia RTX PRO 6000 Blackwell Server Edition GPUs in March 2025, with general availability expected this summer.
The platform, dubbed Cirrascale Inference Platform, will automatically select the best AI accelerator to adapt to changing model requirements, user demands, and workflow shifts.
In October 2024, Cirrascale added the Nvidia HGX H200 servers to its AI Innovation Cloud.
Cirrascale was notably one of the first companies to get access to the Nvidia H100 GPUs. The company was previously known as Verari Technologies, before being renamed in 2010 as Cirrascale. The cloud services provider is a known customer of Digital Realty data centers.
