A Tuning Studio experiment has a single hardware configuration.
The following table shows the hardware configuration that is used when tuning foundation models in a tuning experiment.
| Capacity type | Capacity units per hour |
|---|---|
| NVIDIA A100 80GB GPU | 43 |
Compute usage in projects
Tuning Studio consumes compute resources as CUH from the watsonx.ai Runtime service.
You can monitor the total monthly amount of CUH consumption for the watsonx.ai Runtime service on the Resource usage page on the Manage tab of your project.
What consumes CUH?
Capacity unit hours (CUH) are consumed for running assets, not for working in tools. There is no consumption charge for defining a tuning experiment, but there is a charge for running the experiment. Similarly, there is no charge for defining a deployment or deployment job for a tuned model, but there is a charge for running a deployment job or inferencing against the deployed model.
Learn more
- Tuning Studio
- watsonx.ai Runtime plans
- Compute resource options for assets and deployments in spaces
- Monitoring account resource usage
Parent topic: Choosing compute resources for tools