Run pre-trained models

You can deploy your trained model in Supervisely with just a click or pick one of the ready-to-use pre-trained models from the Ecosystem. Search for “Serve …” Supervisely App and run the model on one of your agents. Now, use API to communicate with the model or use one of our Supervisely Apps that helps you apply the model in various scenarios, such as running the model on each image in the dataset or integrating it in the Labeling Toolbox.

Some examples of pre-trained models:

• Serve MMDetection

• Serve MMClassification

• Serve Segment Anything in High Quality

• Serve UNet

Find more pre-trained models in Ecosystem

Some examples of model applications:

• Apply NN to Images Project

• NN Image Labeling

• Batched Smart Tool

• Apply OWL-ViT To Images Project

• Apply Detection and Pose Estimation Models to Images Project

Guides on running models in Supervisely:

• How to auto-label images with OWL-ViT - SOTA Google's foundation object detector
• ViTPose — How to use the best Pose Estimation Model on Humans & Animals
• Unlocking Text - OCR with Supervisely
• MBPTrack Tutorial - SOTA 3D Point Cloud Object Tracking in 2023 for LiDAR & Radar