Gcore launches GPU virtual machines on NVIDIA Hopper

Key Takeaways

• Gcore launches GPU Virtual Machines (VMs) on NVIDIA Hopper, providing flexible and cost-efficient access to AI compute
• The new VMs offer high-performance computing to a broad range of customers, with dynamic scaling to meet changing infrastructure needs
• Gcore's GPU VMs are available in the Sines-3 sovereign AI region in Portugal, with access to NVIDIA Hopper GPUs and high-bandwidth NVIDIA Quantum InfiniBand networking
• The flexible deployment model is ideal for AI startups, EU R&D labs, and research institutions with variable compute requirements

Introduction to Gcore's GPU Virtual Machines

Gcore has announced the launch of GPU Virtual Machines (VMs) on NVIDIA Hopper, delivering flexible and cost-efficient access to AI compute. As AI development becomes increasingly iterative and central to company functioning, organizations require infrastructure that can scale dynamically to meet their changing needs.

Benefits of Gcore's GPU VMs

The new GPU VMs provide access to high-performance computing, making it more accessible to a broad range of customers. With Gcore's VMs, companies can match GPU capacity and cost to the stage of their project with precision, allowing for more efficient use of resources. The VMs offer the same NVIDIA Hopper GPUs with high-bandwidth NVIDIA Quantum InfiniBand networking as Gcore's Bare Metal GPU Cloud, but without requiring a long-term commitment to hardware.

Comparison of Gcore's GPU VMs and Bare Metal GPU Cloud

Cutting Idle Costs without Complexity

Gcore's GPU VMs allow companies to cut idle costs without complexity, as the GPU billing pauses automatically when the instance is powered off. Volumes, IPs, and configuration remain intact, but the GPU meter stops running, so companies only pay for storage and IPs while paused. When ready to start work again, teams can restart the VMs without needing to set up or reconfigure.

Conclusion

Gcore's launch of GPU Virtual Machines on NVIDIA Hopper provides a flexible and cost-efficient solution for companies requiring high-performance computing for AI workloads. The new VMs offer dynamic scaling, flexible deployment models, and reduced idle costs, making them an ideal choice for AI startups, EU R&D labs, and research institutions.

Bottom Line

Gcore's GPU VMs on NVIDIA Hopper offer a powerful and flexible solution for companies looking to accelerate their AI development without incurring high fixed costs. With its dynamic scaling, flexible deployment models, and reduced idle costs, Gcore's GPU VMs are poised to revolutionize the way companies approach AI compute, providing a more efficient and cost-effective solution for a wide range of use cases.