GPUs

qBraid Lab GPU instances support accelerated circuit simulation and quantum machine learning, including GPU-based quantum gradients.

Available GPU configurations

qBraid offers NVIDIA GPU instances with Blackwell, Hopper, Ampere, and Ada Lovelace architectures:

• NVIDIA B200 (Blackwell) is available in 1x, 2x, 4x, and 8x configurations.
• NVIDIA H200 and H100 (Hopper) are available in configurations up to 8x for large workloads.
• NVIDIA GH200 (Grace Hopper) combines CPU and GPU hardware in a superchip.
• NVIDIA A100 (Ampere) is available in configurations up to 8x.
• NVIDIA L4, L40S, RTX 4090, RTX 5090, and RTX 6000 Ada provide lower-cost options for development and inference.

Billing is in credits/minute with rates shown in your account launcher. See GPU pricing for how to check current rates.

Launch a GPU instance

From your account dashboard, switch to the On-Demand tab. Each GPU profile shows its per-minute credit rate and a real-time availability indicator. Click Launch next to an available profile to start the instance.

Once the instance is running, you can open it in JupyterLab, VS Code, or connect via the built-in terminal, all from the browser. See Accessing Your Instance for details.

Managing GPU instances

GPU instances follow the on-demand lifecycle. Stop and resume are available only when the provider can preserve the instance disk:

• Stop pauses compute billing and preserves your disk and files for resumption.
• Terminate permanently deletes the instance and its data, ending all billing.

See On-Demand Instances for the complete lifecycle guide, including file persistence, billing details, and concurrent instance limits.

GPU utilities

Inspect your GPU hardware with the NVIDIA System Management Interface (nvidia-smi) and NVIDIA CUDA Toolkit (nvcc) command line utilities.