Lambda, which operates a GPU cloud powered by NVIDIA GPUs, has launched a new service, Lambda 1-Click Clusters. This service provides AI engineers and researchers with short-term access to multi-node GPU clusters in the cloud for large-scale AI model training. It marks the first time such access to NVIDIA H100 Tensor Core GPUs on 2 to 64 nodes has been made available on demand through a self-serve cloud service, without the need for expensive long-term contracts. “It has been our best experience in training runs, and as a team they have been super responsive and supportive across the board,” said Mahmoud Felfel, co-founder of PlayHT.
Lambda 1-Click Clusters are designed to meet the specific needs of today’s AI teams, who may not require continuous access to top-end GPUs. These teams can now quickly spin up a short-term cluster with hundreds of GPUs for a few weeks to run experiments, pause without wasting idle GPU time, and prepare for the next iteration. This flexibility ensures that AI innovation is not hampered by financial or contractual limitations. Robert Brooks, founding team member and VP of Revenue at Lambda, highlighted the significance of this launch: “Lambda has solved a complex compute challenge only a few very large companies have: partitioning a large, high-performant AI deployment to make smaller GPU clusters.”
Key Points:
• Service Name: Lambda 1-Click Clusters
• Target Users: AI engineers and researchers
• Hardware: NVIDIA H100 Tensor Core GPUs, NVIDIA Quantum-2 InfiniBand networking
• Cluster Size: 2 to 64 nodes, 16 to 512 GPUs
• Minimum Reservation: Two weeks
• Self-Serve Model: On-demand access without long-term contracts
• Recent Developments: $500 million GPU-backed facility, $320 million Series C funding round
• User Feedback: Praised for ease of setup, stable infrastructure, and responsive support