Inference service configurations

Last update: October 13 2026Open in Markdown

When creating an inference service, you can select its configuration. The list of available configurations depends on the selected model and its parameters. The configurations automatically select the number and type of graphics processing units (GPUs), number of vCPUs, and RAM.

Each configuration card displays expected model performance indicators. You can compare configurations and choose the one required for your tasks.

After creating an inference service, the configuration cannot be changed.