Cirrascale com

Cirrascale provides cloud-based AI infrastructure to accelerate private AI training and inference workloads. According to its official information, the service is positioned as an "Expert Neocloud," aimed at enterprises and institutions that need private AI capabilities.

1. Main Uses Cirrascale's cloud solution is designed to support private AI training and inference. The platform emphasizes high throughput, multi-tier storage, high-bandwidth low-latency networking, and multi-GPU server and storage solutions to help users run AI projects.

2. Supported Accelerators Official materials show that its cloud infrastructure supports multiple leading accelerator brands such as NVIDIA, AMD, and Qualcomm, and allows users to test and deploy across various accelerators within the same cloud environment.

3. Main Capabilities and Features

• No egress/ingress data transfer fees.
• High-bandwidth, low-latency networking.
• Multi-tier storage and higher throughput.
• Provides professional and managed services to achieve zero DevOps.
• Customized server and storage solutions for multiple GPUs.
• Supports testing and deploying multiple accelerators on the same cloud platform.

1. Collaboration Solutions with Google Cirrascale collaborates with Google to provide secure, private AI solutions for enterprises and the public sector, specifically including:

• Private Gemini on Google Distributed Cloud: Run Gemini on Google Distributed Cloud through the Cirrascale inference platform, keeping data behind the firewall.
• Google GPAR Services: In partnership with Google Public Sector, provide secure, high-performance AI infrastructure for research institutions and universities to accelerate scientific discovery.
• Private Gemini Endpoint on Google Cloud Marketplace: Cirrascale's private Gemini endpoint is available on Google Cloud Marketplace, keeping data within the user's own infrastructure and providing routing, visibility, and control capabilities.

1. Applicable Scenarios According to the official description, the service is suitable for enterprises, the public sector, research institutions, and universities that need private AI training and inference. Its managed services can help teams lacking DevOps resources reduce operational burden.

2. Limitations and Missing Information Currently, official materials do not provide specific details such as pricing, billing models, service level agreements (SLAs), available regions, free trials, or registration steps. This information is not yet available.