Together AI announced the general availability of its Together Instant Clusters service, which automates the provisioning of GPU clusters.

The “pay-as-you-go” service lets organizations quickly spin up NVIDIA GPU clusters, ranging from a single node with eight GPUs to large, multi-node systems with hundreds of processors. “Each cluster is built using NVIDIA H100 (80GB SXM) GPUs, engineered for AI training, inference, and fine-tuning at scale,” the company says. Instant Clusters also feature NVIDIA Quantum-2 InfiniBand and NVLink networking, with the choice of Kubernetes or Slurm for orchestration.

According to the company, Instant Clusters can be provisioned in minutes and include the following components:

• GPU Operator to manage drivers and runtime software.
• Ingress controller to handle traffic into your cluster.
• NVIDIA Network Operator for high-performance networking.
• Cert manager for secure certificates and HTTPS endpoints.

See pricing and other details at Together AI.