SAM 3

v3

Class: SegmentAnything3BlockV3 (there are multiple versions of this block)

Source: inference.core.workflows.core_steps.models.foundation.segment_anything3.v3.SegmentAnything3BlockV3

Warning: This block has multiple versions. Please refer to the specific version for details. You can learn more about how versions work here: Versioning

Run Segment Anything 3 (SAM3), a zero-shot instance segmentation model, on an image.

You can use text prompts for open-vocabulary segmentation - just specify class names and SAM3 will segment those objects in the image.

This block supports two output formats: - rle (default): Returns masks in RLE (Run-Length Encoding) format, which is more memory-efficient - polygons: Returns polygon coordinates for each mask

RLE format is recommended for high-resolution images or workflows with many detections.

Type identifier

Use the following identifier in step "type" field: roboflow_core/sam3@v3to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
model_id	str	model version. You only need to change this for fine tuned sam3 models..	✅
class_names	Optional[List[str], str]	List of classes to recognise.	✅
confidence	float	Minimum confidence threshold for predicted masks.	✅
per_class_confidence	List[float]	List of confidence thresholds per class (must match class_names length).	✅
apply_nms	bool	Whether to apply Non-Maximum Suppression across prompts.	✅
nms_iou_threshold	float	IoU threshold for cross-prompt NMS. Must be in [0.0, 1.0].	✅
output_format	str	'rle' returns efficient RLE encoding (recommended), 'polygons' returns polygon coordinates.	❌

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 SAM 3 in version v3.

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

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds SAM 3 in version v3 has.

Bindings

• input

  • images (image): The image to infer on..
  • model_id (roboflow_model_id): model version. You only need to change this for fine tuned sam3 models..
  • class_names (Union[string, list_of_values]): List of classes to recognise.
  • confidence (float): Minimum confidence threshold for predicted masks.
  • per_class_confidence (list_of_values): List of confidence thresholds per class (must match class_names length).
  • apply_nms (boolean): Whether to apply Non-Maximum Suppression across prompts.
  • nms_iou_threshold (float): IoU threshold for cross-prompt NMS. Must be in [0.0, 1.0].

• output

  • predictions (Union[rle_instance_segmentation_prediction, instance_segmentation_prediction]): Prediction with detected bounding boxes and RLE-encoded segmentation masks in form of sv.Detections(...) object if rle_instance_segmentation_prediction or Prediction with detected bounding boxes and segmentation masks in form of sv.Detections(...) object if instance_segmentation_prediction.

Example JSON definition of step SAM 3 in version v3

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/sam3@v3",
    "images": "$inputs.image",
    "model_id": "sam3/sam3_final",
    "class_names": [
        "car",
        "person"
    ],
    "confidence": 0.3,
    "per_class_confidence": [
        0.3,
        0.5,
        0.7
    ],
    "apply_nms": "<block_does_not_provide_example>",
    "nms_iou_threshold": 0.5,
    "output_format": "rle"
}

v2

Class: SegmentAnything3BlockV2 (there are multiple versions of this block)

Source: inference.core.workflows.core_steps.models.foundation.segment_anything3.v2.SegmentAnything3BlockV2

Warning: This block has multiple versions. Please refer to the specific version for details. You can learn more about how versions work here: Versioning

Run Segment Anything 3, a zero-shot instance segmentation model, on an image.

You can pass in boxes/predictions from other models as prompts, or use a text prompt for open-vocabulary segmentation. If you pass in box detections from another model, the class names of the boxes will be forwarded to the predicted masks.

Type identifier

Use the following identifier in step "type" field: roboflow_core/sam3@v2to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
model_id	str	model version. You only need to change this for fine tuned sam3 models..	✅
class_names	Optional[List[str], str]	List of classes to recognise.	✅
confidence	float	Minimum confidence threshold for predicted masks.	✅
per_class_confidence	List[float]	List of confidence thresholds per class (must match class_names length).	✅
apply_nms	bool	Whether to apply Non-Maximum Suppression across prompts.	✅
nms_iou_threshold	float	IoU threshold for cross-prompt NMS. Must be in [0.0, 1.0].	✅

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 SAM 3 in version v2.

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

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds SAM 3 in version v2 has.

Bindings

• input

  • images (image): The image to infer on..
  • model_id (roboflow_model_id): model version. You only need to change this for fine tuned sam3 models..
  • class_names (Union[string, list_of_values]): List of classes to recognise.
  • confidence (float): Minimum confidence threshold for predicted masks.
  • per_class_confidence (list_of_values): List of confidence thresholds per class (must match class_names length).
  • apply_nms (boolean): Whether to apply Non-Maximum Suppression across prompts.
  • nms_iou_threshold (float): IoU threshold for cross-prompt NMS. Must be in [0.0, 1.0].

• output

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

Example JSON definition of step SAM 3 in version v2

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/sam3@v2",
    "images": "$inputs.image",
    "model_id": "sam3/sam3_final",
    "class_names": [
        "car",
        "person"
    ],
    "confidence": 0.3,
    "per_class_confidence": [
        0.3,
        0.5,
        0.7
    ],
    "apply_nms": "<block_does_not_provide_example>",
    "nms_iou_threshold": 0.5
}

v1

Class: SegmentAnything3BlockV1 (there are multiple versions of this block)

Source: inference.core.workflows.core_steps.models.foundation.segment_anything3.v1.SegmentAnything3BlockV1

Warning: This block has multiple versions. Please refer to the specific version for details. You can learn more about how versions work here: Versioning

Run Segment Anything 3, a zero-shot instance segmentation model, on an image.

You can pass in boxes/predictions from other models as prompts, or use a text prompt for open-vocabulary segmentation. If you pass in box detections from another model, the class names of the boxes will be forwarded to the predicted masks.

Type identifier

Use the following identifier in step "type" field: roboflow_core/sam3@v1to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
model_id	str	model version. You only need to change this for fine tuned sam3 models..	✅
class_names	Optional[List[str], str]	List of classes to recognise.	✅
threshold	float	Threshold for predicted mask scores.	✅

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 SAM 3 in version v1.

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

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds SAM 3 in version v1 has.

Bindings

• input

  • images (image): The image to infer on..
  • model_id (roboflow_model_id): model version. You only need to change this for fine tuned sam3 models..
  • class_names (Union[string, list_of_values]): List of classes to recognise.
  • threshold (float): Threshold for predicted mask scores.

• output

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

Example JSON definition of step SAM 3 in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/sam3@v1",
    "images": "$inputs.image",
    "model_id": "sam3/sam3_final",
    "class_names": [
        "car",
        "person"
    ],
    "threshold": 0.3
}