Segment Anything 2 Model

Class: SegmentAnything2BlockV1

Source: inference.core.workflows.core_steps.models.foundation.segment_anything2.v1.SegmentAnything2BlockV1

Run Segment Anything 2, a zero-shot instance segmentation model, on an image.

** Dedicated inference server required (GPU recomended) **

You can use pass in boxes/predictions from other models to Segment Anything 2 to use as prompts for the model. If you pass in box detections from another model, the class names of the boxes will be forwarded to the predicted masks. If using the model unprompted, the model will assign integers as class names / ids.

Type identifier

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

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
version	str	Model to be used. One of hiera_large, hiera_small, hiera_tiny, hiera_b_plus.	✅
threshold	float	Threshold for predicted masks scores.	✅
multimask_output	bool	Flag to determine whether to use sam2 internal multimask or single mask mode. For ambiguous prompts setting to True is recomended..	✅

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 Segment Anything 2 Model in version v1.

• inputs: Mask Visualization, Classification Label Visualization, Detections Consensus, Detections Merge, Instance Segmentation Model, Webhook Sink, Multi-Label Classification Model, Email Notification, QR Code Generator, VLM As Detector, LMM, SAM 3, Detection Offset, Corner Visualization, Image Convert Grayscale, Stability AI Outpainting, Segment Anything 2 Model, Halo Visualization, Object Detection Model, JSON Parser, Trace Visualization, Google Vision OCR, Instance Segmentation Model, CSV Formatter, Text Display, Stitch Images, Google Gemini, Local File Sink, Slack Notification, VLM As Classifier, PTZ Tracking (ONVIF).md), Roboflow Dataset Upload, Color Visualization, Dot Visualization, Polygon Visualization, Object Detection Model, Anthropic Claude, Byte Tracker, Contrast Equalization, Identify Changes, Detections Classes Replacement, Velocity, Moondream2, SIFT Comparison, Halo Visualization, Florence-2 Model, Blur Visualization, Label Visualization, Twilio SMS/MMS Notification, Ellipse Visualization, OpenAI, SIFT, Model Monitoring Inference Aggregator, Single-Label Classification Model, Detections List Roll-Up, OpenAI, Image Threshold, Background Color Visualization, Model Comparison Visualization, OpenAI, Keypoint Detection Model, Gaze Detection, Polygon Visualization, SAM 3, Twilio SMS Notification, Bounding Box Visualization, OCR Model, Overlap Filter, Icon Visualization, Time in Zone, Google Gemini, Florence-2 Model, Roboflow Dataset Upload, Anthropic Claude, Dynamic Zone, Dynamic Crop, VLM As Detector, Google Gemini, Path Deviation, Image Blur, Line Counter, Byte Tracker, Stability AI Inpainting, Template Matching, Image Contours, Path Deviation, Morphological Transformation, Triangle Visualization, Bounding Rectangle, Detections Stitch, Relative Static Crop, Detections Filter, Camera Calibration, Grid Visualization, Detections Stabilizer, Camera Focus, Image Slicer, Detections Combine, LMM For Classification, Line Counter Visualization, Keypoint Detection Model, Llama 3.2 Vision, SIFT Comparison, Camera Focus, Time in Zone, Background Subtraction, Image Slicer, Circle Visualization, Seg Preview, Identify Outliers, Clip Comparison, Email Notification, Byte Tracker, Image Preprocessing, SAM 3, Depth Estimation, Cosine Similarity, Time in Zone, CogVLM, Absolute Static Crop, Roboflow Custom Metadata, EasyOCR, Stitch OCR Detections, Perspective Correction, Anthropic Claude, Pixelate Visualization, Stability AI Image Generation, Reference Path Visualization, Keypoint Visualization, VLM As Classifier, Detection Event Log, Polygon Zone Visualization, YOLO-World Model, Stitch OCR Detections, Crop Visualization, Motion Detection, OpenAI, Detections Transformation
• outputs: Mask Visualization, Circle Visualization, Detections Consensus, Detections Merge, Halo Visualization, Dynamic Zone, Florence-2 Model, Blur Visualization, Florence-2 Model, Dynamic Crop, Label Visualization, Path Deviation, Detection Offset, Corner Visualization, Ellipse Visualization, Byte Tracker, Byte Tracker, Line Counter, Model Monitoring Inference Aggregator, Segment Anything 2 Model, Halo Visualization, Stability AI Inpainting, Detections List Roll-Up, Model Comparison Visualization, Background Color Visualization, Size Measurement, Trace Visualization, Path Deviation, Time in Zone, Line Counter, Triangle Visualization, Bounding Rectangle, Detections Stitch, Detections Filter, Roboflow Custom Metadata, Detections Stabilizer, Roboflow Dataset Upload, PTZ Tracking (ONVIF).md), Camera Focus, Perspective Correction, Color Visualization, Detections Combine, Pixelate Visualization, Polygon Visualization, Dot Visualization, Polygon Visualization, Bounding Box Visualization, Detection Event Log, Byte Tracker, Distance Measurement, Detections Classes Replacement, Overlap Filter, Icon Visualization, Crop Visualization, Time in Zone, Time in Zone, Velocity, Detections Transformation, Roboflow Dataset Upload

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Segment Anything 2 Model in version v1 has.

Bindings

• input

  • images (image): The image to infer on..
  • boxes (Union[instance_segmentation_prediction, object_detection_prediction, keypoint_detection_prediction]): Bounding boxes (from another model) to convert to polygons.
  • version (string): Model to be used. One of hiera_large, hiera_small, hiera_tiny, hiera_b_plus.
  • threshold (float): Threshold for predicted masks scores.
  • multimask_output (boolean): Flag to determine whether to use sam2 internal multimask or single mask mode. For ambiguous prompts setting to True is recomended..

• output

  • predictions (instance_segmentation_prediction): Prediction with detected bounding boxes and segmentation masks in form of sv.Detections(...) object.

Example JSON definition of step Segment Anything 2 Model in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/segment_anything@v1",
    "images": "$inputs.image",
    "boxes": "$steps.object_detection_model.predictions",
    "version": "hiera_large",
    "threshold": 0.3,
    "multimask_output": true
}