> For the complete documentation index, see [llms.txt](https://docs.roboflow.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.roboflow.com/datasets/annotate/annotate/ai-labeling/model-assisted-labeling.md).
# Label Assist
Reduce annotation time and get a model into production faster by using model weights from a trained model to label your data
{% hint style="info" %}
Label Assist is one of many [AI Labeling](/datasets/annotate/annotate/ai-labeling.md) features. Using this feature will consume [credits](https://docs.roboflow.com/platform/billing-and-plans/credits) at the rates listed on our [credits page](https://roboflow.com/credits).
{% endhint %}
Label Assist allows you to use a foundation model, a previous version of a trained model, or a public model on Roboflow Universe as an annotation assistant. Model-assisted labeling is ideal for identifying specific classes in an image automatically.
When model-assisted labeling is enabled, the selected model will run when you open an image in the annotation tool. The model will predict objects in the image. Annotations are added for each object the model finds.
## How to Use Label Assist
Open an image in Roboflow Annotate then click the magic wand icon in the command tray.
{% hint style="info" %}
Label Assist is currently supported on select project types. If there's a project type you'd like to see supported, let us know!
{% endhint %}
A pop-up will appear asking what model you want to use to help you label images:
Select a model from the pop up. The pop up opens on the "AI Models" tab, which labels the open image with a foundation model. You can select a model to which you have access through the `Your Models` tab. You can select any model from Universe through the `Public Models` tab. You need to star a model on Universe before it will show up in the `Public Models` tab.
The steps below cover the `Your Models` and `Public Models` tabs. To use a foundation model instead, see [AI Models](#ai-models).
Click "Continue" once you have selected a model.
Next, you will be asked what classes you want to find. By default, the selected model will return predictions for all classes. You can uncheck classes the model should return from the pop up.
If you want to remap any class names, you can do so. This is useful if the model you are using has different class names from the classes you want to add to images in your dataset.
On a segmentation project, the same pop up has a "Mask output" choice. Label Assist returns polygons by default. Pick the pixel mask (RLE) option to get masks instead, which keeps separate regions and holes that a polygon drops. Roboflow remembers your choice.
Once you have configured the classes for use with the selected model, click the "Select Classes" button. You will then be able to start Label Assist.
Label Assist will start running on the current image you are annotating. Label Assist will persist as you navigate through images in your dataset using the forward and back arrows in the annotation interface.
Here is an example of Label Assist recommending an annotation:
## AI Models
The "AI Models" tab labels the image you have open with a foundation model, so you can label with AI before you train a model. It uses the same models as [Auto Label](/datasets/annotate/annotate/ai-labeling/auto-label.md), and it draws boxes only.
Pick a model from the "Model" list. Gemini 4 Argon is selected for you. A model your Workspace cannot run is greyed out with the reason, and Label Assist opens on the `Your Models` tab when no AI model is available.
Click the classes you want to find. Click "+ Add class" to add a class to your project without leaving the editor. Each class you pick gets a description box. The model searches for the description and saves what it finds under the class name, so the description "green umbrella" on the class "umbrella" adds boxes labeled "umbrella". Leave the box empty to search for the class name.
Click "Label with AI" to run the model on the image. The boxes come back as suggestions, so nothing is saved yet. Drag a box to move it, drag a corner to resize it, pick a different class in the "Class" list, or click "Delete selected box". Use the arrows to step through the detections one at a time.
The add button counts the boxes that are left: click it to save them as annotations, or click "Discard suggestions" to drop all of them. Annotations already on the image stay as they are.
