Perception Encoder Embedding Model

Class: PerceptionEncoderModelBlockV1

Source: inference.core.workflows.core_steps.models.foundation.perception_encoder.v1.PerceptionEncoderModelBlockV1

Use the Meta Perception Encoder model to create semantic embeddings of text and images.

This block accepts an image or string and returns an embedding. The embedding can be used to compare similarity between different images or between images and text.

Type identifier

Use the following identifier in step "type" field: roboflow_core/perception_encoder@v1to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Unique name of step in workflows.	❌
data	str	The string or image to generate an embedding for..	✅
version	str	Variant of Perception Encoder model.	✅

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Available Connections

Compatible Blocks

Check what blocks you can connect to Perception Encoder Embedding Model in version v1.

• inputs: Triangle Visualization, Local File Sink, VLM as Classifier, Slack Notification, Twilio SMS Notification, Google Vision OCR, SIFT Comparison, OCR Model, CSV Formatter, OpenAI, Email Notification, Instance Segmentation Model, Camera Calibration, Ellipse Visualization, Model Comparison Visualization, Corner Visualization, Pixelate Visualization, LMM For Classification, Reference Path Visualization, Roboflow Custom Metadata, Keypoint Detection Model, OpenAI, Mask Visualization, Model Monitoring Inference Aggregator, Label Visualization, SIFT, Image Convert Grayscale, Stability AI Outpainting, Polygon Zone Visualization, Stability AI Inpainting, Llama 3.2 Vision, Perspective Correction, Image Preprocessing, Clip Comparison, Dot Visualization, Camera Focus, Depth Estimation, OpenAI, Polygon Visualization, Keypoint Visualization, Image Blur, Image Slicer, Image Contours, Florence-2 Model, CogVLM, Google Gemini, Background Color Visualization, Roboflow Dataset Upload, Dynamic Crop, Halo Visualization, Classification Label Visualization, Stitch OCR Detections, Single-Label Classification Model, Crop Visualization, VLM as Detector, Anthropic Claude, Blur Visualization, Color Visualization, Relative Static Crop, Object Detection Model, Bounding Box Visualization, Webhook Sink, Grid Visualization, Stitch Images, Multi-Label Classification Model, Image Slicer, Line Counter Visualization, Florence-2 Model, Stability AI Image Generation, Image Threshold, Trace Visualization, Roboflow Dataset Upload, Absolute Static Crop, Circle Visualization, LMM
• outputs: Identify Changes, Identify Outliers, Cosine Similarity

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Perception Encoder Embedding Model in version v1 has.

Bindings

• input

  • data (Union[string, image]): The string or image to generate an embedding for..
  • version (string): Variant of Perception Encoder model.

• output

  • embedding (embedding): A list of floating point numbers representing a vector embedding..

Example JSON definition of step Perception Encoder Embedding Model in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/perception_encoder@v1",
    "data": "$inputs.image",
    "version": "PE-Core-B16-224"
}