Dedicated Deployments

A Dedicated Deployment is a remote server managed by Roboflow on which you can run computer vision models supported by the Roboflow Inference server. This includes object detection, segmentation, classification, and keypoint models trained on or uploaded to Roboflow, as well as foundation models like CLIP.

Dedicated Deployments allow you to have cloud servers allocated specifically for your use.

Dedicated Deployments are accessible with your workspace's API key. You can send inference requests to a Dedicated Deployment as if it is running locally.

Dedicated Deployment Types

There are two types of Dedicated Deployments:

• Dev CPU (A CPU machine with no GPU)

• Dev GPU

All servers will run Roboflow Inference, our on-device inference server. Review the Roboflow Inference documentation to learn more about all of the features available.

Roboflow Inference automatically uses GPUs in your Dedicated Deployment if you chose a GPU-enabled deployment type.

Provision and Manage Dedicated Deployments (Web Application)

You can provision, manage, and delete Dedicated Deployments in the Roboflow Workflows web application.

Roboflow Workflows is a low-code, web-based application builder for creating computer vision applications.

To create a Dedicated Deployment, first create a Roboflow Workflow. To do so, click on Workflows on the left tab in the Roboflow dashboard, then click "Create Workflow":

Then, click on the "Running on Hosted API" link in the top left corner:

Click Dedicated Deployments to create and see your Dedicated Deployments:

Set a name for your Deployment, then choose whether you need a CPU or GPU.

Then, click "Create Dedicated Deployment".

Your Deployment will be provisioned. It may take anywhere from a few seconds to a few minutes to provision your deployment.

When your Deployment is ready, the status will be updated to Ready. You can then click "Connect" to use your Deployment with your Workflow in the Workflows editor:

Manage Dedicated Deployments (CLI)

Dedicated deployments are managed by using the Roboflow Command Line Interface (CLI). The minimum version supporting this feature is 1.1.41. You can install the CLI using:

pip install roboflow

Once you have installed the CLI, authenticate with:

roboflow login

Create a Dedicated Deployment

To provision a Dedicated Deployment, run:

roboflow deployment add DEPLOYMENT_NAME -m MACHINE_TYPE -t DURATION -w WAIT_ON_PENDING -e NO_DELETE_ON_EXPIRATION -v INFERENCE_VERSION

Above, replace:

• DEPLOYMENT_NAME: deployment name, 5~15 lower case alphanumeric characters and dash (a-z, 0-9, -), the first character must be a letter (a-z), must be unique within your workspace.

• MACHINE_TYPE: machine type, run roboflow deployment machine_type to get available options.

• DURATION: how long you want this deployment to be active (unit: hour), must be within 0.1~6, the default value is 3.

• WAIT_ON_PENDING: wait for the deployment to be ready (or failed), default is false.

• NO_DELETE_ON_EXPIRATION: not delete the deployment when expired, default is false.

• INFERENCE_VERSION: version of the inference server docker image, must be 0.16.0 or above. See available options at docker hub for cpu and gpu deployments.

List Dedicated Deployments

To list all Dedicated Deployments associated with your Roboflow Workspace, run:

roboflow deployment list

Check Dedicated Deployment Status

To check the status of a Dedicated Deployment, run

roboflow deployment get DEPLOYMENT_NAME -w WAIT_ON_PENDING

Above, set:

• DEPLOYMENT_NAME: deployment name.

• WAIT_ON_PENDING: wait for the deployment to be ready (or failed), default is false.

This command will return information such as:

• The URL associated with your Dedicated Deployment;

• The status of the Deployment, and;

• The instance type.

Delete a Dedicated Deployment

You can delete a Dedicated Deployment at any time. You cannot recover deleted deployments.

To delete a Deployment, run:

roboflow deployment delete DEPLOYMENT_NAME

Above, replace:

• DEPLOYMENT_NAME: deployment name.