Create an inference service

1. Select a model

In the control panel, in the top menu, click Products and select Dedicated Inference.
Select a location. Once the inference service is created, the location cannot be changed.
In the model card, click Create.
Enter the name of the inference service.
To filter inference services in the list, add tags. A tag with the model name is added automatically. To add a new tag, in the Tags field, enter the tag and press Enter.
Optional: enter a description for the inference service. For example, specify its purpose.
Click Continue.

Specify model parameters.

1.1. Select the data type for the KV cache.

1.2. Select the maximum context length.
Select an inference service configuration. When selecting, consider the expected model performance indicators.

You cannot change the configuration after creating the inference service.
Click Continue.

Configure the number of inference instances.

1.1. To have a fixed number of instances in the service, open the Fixed tab and specify the number of instances.

1.2. To use autoscaling in the service, open the With autoscaling tab and set the minimum and maximum number of instances. The number of instances will change automatically only within the specified range depending on the load on the inference service.

You can change the number of inference instances after creating the inference service. For details, see the guide Scale inference service.
Select the disk type for the inference instance.
Select an inference service connection type:
- public — access via a public endpoint from the internet;
- private — access via a private endpoint only from devices in the inference service's private subnet.
You cannot change the connection type after creating the inference service.
Click Continue.

Review the final inference service configuration.
Check the inference service price.
Click Create Inference Service. Creating an inference service may take about 15 minutes.