AI Compute Portfolio Built for Every Scale, from Local Development to Production Deployment

CORSAIR^® announced the launch of CORSAIR PRO, a new portfolio of AI workstations and servers designed to support and scale with the rapidly evolving needs of modern AI computing. This marks an important expansion of CORSAIR’s strategy, extending the Company’s performance engineering expertise into purpose-built AI compute infrastructure.

Key Highlights:

• Features NVIDIA Grace Blackwell-based systems, including the FlexPrime V80B workstation powered by the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip
• Expands Corsair offerings in AI infrastructure solutions.
• Built for modern AI workloads, including model development, fine-tuning, inference, agentic AI, computer vision, and production deployment.
• Scales from developer workstations to server-class platforms for shared compute and high-density AI workloads.
• Optional Deployment software stacks to tune each system configuration.

“We’re introducing CORSAIR PRO to address a real shift in the market,” said Matthew Hsu, SVP and GM of CORSAIR Components and Systems. “AI teams need infrastructure that matches how they actually work, whether that starts with a local workstation or for the data center application.”

The new CORSAIR PRO portfolio includes AI Workstations built for development, prototyping, fine-tuning, and AI execution. This lineup includes the FlexPrime V20R, FlexPrime V50R, FlexPrime V80T, and FlexPrime R80T, giving customers a range of platforms suited to individual developers, research labs, and engineering teams running shared training and inference workloads.

CORSAIR PRO also introduces a family of AI Servers built for larger-scale training, inference, and deployment. The server lineup includes the FlexGrid G2E2, FlexGrid MG4E2, FlexGrid HG8E2, and FlexGrid MI8E2, covering a broad set of requirements across entry AI infrastructure, balanced multi-GPU compute, high-density training, and large-memory accelerator environments.

Together, these systems are designed to give customers the flexibility to choose infrastructure based on where they are in the AI workflow today, while also giving them room to scale as requirements grow.

Among the systems being highlighted is the FlexPrime V80B, a new CORSAIR PRO workstation designed for demanding AI training, fine-tuning, inference, and simulation workloads. Powered by the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip, the FlexPrime V80B brings advanced shared-memory architecture to a deskside platform, helping users work more effectively with large models and memory-intensive datasets.

CORSAIR PRO is designed to support real-world AI workflows, including training, fine-tuning, inference, RAG pipelines, AI assistants, agentic workflows, and computer vision. By offering both workstation and server platforms, CORSAIR PRO gives customers options for the full AI lifecycle, from early development through deployment.

CORSAIR PRO AI systems can be configured with a validated software stack tuned to the hardware in each build. PyTorch and TensorFlow are pre-installed and validated against the selected GPUs. Docker and Kubernetes are configured for reproducible environments and multi-node deployment. GPU drivers, CUDA, and ROCm versions are matched to the hardware before the system ships. Customers who select the stack spend their first day training models, not configuring environments.

“AI is reshaping every industry, and CORSAIR PRO enables Corsair to participate in that growth targeting local, private and secure compute that is easy to deploy, and can built to scale,” said Thi La, Chief Executive Officer of Corsair. “This expansion moves CORSAIR into professional AI infrastructure, broadens our customer base, and positions us to capture higher-value system opportunities in AI compute.”

Source: BUSINESS WIRE