TensorFlow.js models

Explore pre-trained models to add computer vision, natural language processing (NLP), and other common ML tasks to your web and browser-based applications.


Analyze features in images and videos. Unlock new real-time experiences in the browser.

Image classification

Classify images with labels from the ImageNet database (MobileNet).

Object detection

Localize and identify multiple objects in a single image (Coco SSD).

Semantic segmentation

Run semantic segmentation in the browser (DeepLab).


Detect key points and poses on the face, hands, and body with models from MediaPipe and beyond, optimized for JavaScript and Node.js.

Simple face detection

Detect faces in images using a Single Shot Detector architecture with a custom encoder (Blazeface).

Face landmark detection

Predict 486 3D facial landmarks to infer the approximate surface geometry of human faces.

Pose detection

Unified pose detection API for using one of three models that help detect atypical poses and fast body motions with real time performance.

Body segmentation

Segment person(s) and body parts in real-time.

Hand pose detection

Palm detector and hand-skeleton finger tracking model. Predict 21 3D hand keypoints per detected hand.

Portrait depth estimation

Estimate a depth map for a single portrait image of a human.


Enable NLP in your web app using the power of BERT and other Transformer encoder architectures.

Natural language question answering

Answer questions based on the content of a given passage of text using BERT.

Text toxicity detection

Score the perceived impact a comment may have on a conversation, from "Very toxic" to "Very healthy" (Toxicity).

Universal sentence encoder

Encode text into embeddings for NLP tasks such as sentiment classification and textual similarity (Universal Sentence Encoder).


Classify audio to detect sounds and trigger an action in your web app.

Speech command recognition

Classify 1-second audio snippets from the speech commands dataset (speech-commands).


Find more TensorFlow.js models that can be used out of the box.

KNN Classifier

Utility to create a classifier using the K-Nearest-Neighbors algorithm. Can be used for transfer learning.