Object Detection Model¶
v3¶
Class: RoboflowObjectDetectionModelBlockV3 (there are multiple versions of this block)
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 inference on a object-detection model hosted on or uploaded to Roboflow.
You can query any model that is private to your account, or any public model available on Roboflow Universe.
You will need to set your Roboflow API key in your Inference environment to use this block. To learn more about setting your Roboflow API key, refer to the Inference documentation.
Type identifier¶
Use the following identifier in step "type" field: roboflow_core/roboflow_object_detection_model@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 |
Roboflow model identifier.. | ✅ |
confidence_mode |
str |
How confidence thresholds are determined.. | ✅ |
custom_confidence |
float |
Custom confidence threshold for predictions.. | ✅ |
class_filter |
List[str] |
List of accepted classes. Classes must exist in the model's training set.. | ✅ |
iou_threshold |
float |
Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more.. | ✅ |
max_detections |
int |
Maximum number of detections to return.. | ✅ |
class_agnostic_nms |
bool |
Boolean flag to specify if NMS is to be used in class-agnostic mode.. | ✅ |
max_candidates |
int |
Maximum number of candidates as NMS input to be taken into account.. | ✅ |
disable_active_learning |
bool |
Boolean flag to disable project-level active learning for this block.. | ✅ |
active_learning_target_dataset |
str |
Target dataset for active learning, if enabled.. | ✅ |
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 Object Detection Model in version v3.
- inputs:
Absolute Static Crop,VLM As Classifier,Qwen-VL,Triangle Visualization,EasyOCR,Model Comparison Visualization,Detection Event Log,Stitch OCR Detections,Llama 3.2 Vision,Heatmap Visualization,Stitch Images,Keypoint Visualization,Anthropic Claude,Webhook Sink,Google Gemini,GLM-OCR,QR Code Generator,Instance Segmentation Model,VLM As Detector,MoonshotAI Kimi,Multi-Label Classification Model,Twilio SMS/MMS Notification,PLC ModbusTCP,Line Counter,Distance Measurement,Ellipse Visualization,Single-Label Classification Model,Keypoint Detection Model,Pixelate Visualization,Corner Visualization,Buffer,Template Matching,Llama 3.2 Vision,JSON Parser,Qwen 3.5 API,LMM,Roboflow Visual Search,Semantic Segmentation Model,OpenAI-Compatible LLM,Morphological Transformation,Color Visualization,Roboflow Dataset Upload,Dimension Collapse,Morphological Transformation,Camera Calibration,Motion Detection,Google Vision OCR,Clip Comparison,Bounding Box Visualization,OpenAI,Halo Visualization,Object Detection Model,Identify Changes,Contrast Equalization,CSV Formatter,Instance Segmentation Model,Model Monitoring Inference Aggregator,Multi-Label Classification Model,MQTT Writer,PLC Reader,Image Threshold,Blur Visualization,Line Counter Visualization,Google Gemini,Microsoft SQL Server Sink,Single-Label Classification Model,Background Subtraction,VLM As Classifier,Object Detection Model,Identify Outliers,Instance Segmentation Model,Twilio SMS Notification,Polygon Zone Visualization,OPC UA Writer Sink,Google Gemma API,Keypoint Detection Model,Reference Path Visualization,Image Stack,Stability AI Inpainting,Multi-Label Classification Model,Crop Visualization,Anthropic Claude,Image Convert Grayscale,Dynamic Zone,Camera Focus,Image Slicer,Florence-2 Model,Polygon Visualization,MoonshotAI Kimi,Dynamic Crop,Qwen 3.6 API,Event Writer,Perspective Correction,Line Counter,Polygon Visualization,Clip Comparison,OCR Model,Roboflow Visual Search Classifier,Circle Visualization,Slack Notification,CogVLM,Contrast Enhancement,PLC EthernetIP,Grid Visualization,LMM For Classification,Florence-2 Model,OpenAI,Local File Sink,Roboflow Custom Metadata,Qwen3.5-VL,Image Blur,Label Visualization,Object Detection Model,Anthropic Claude,Pixel Color Count,Roboflow Vision Events,Image Preprocessing,Trace Visualization,Email Notification,SIFT Comparison,Stability AI Outpainting,Detections Consensus,PTZ Tracking (ONVIF),Dot Visualization,Current Time,Keypoint Detection Model,Roboflow Asset Library Attributes,Halo Visualization,Classification Label Visualization,Image Contours,Camera Focus,Single-Label Classification Model,OpenRouter,S3 Sink,Stability AI Image Generation,VLM As Detector,Google Gemma,Relative Static Crop,Roboflow Dataset Upload,OpenAI,Image Slicer,Stitch OCR Detections,Google Gemini,Email Notification,Text Display,Semantic Segmentation Model,Size Measurement,Depth Estimation,Icon Visualization,OpenAI,GeoTag Detection,PLC Writer,SIFT Comparison,SIFT,Instance Segmentation Model,Background Color Visualization,Detections List Roll-Up,Mask Visualization - outputs:
Path Deviation,Detections Filter,SAM 3,Instance Segmentation Model,Keypoint Detection Model,Qwen-VL,SAM 3 Interactive,Qwen3-VL,Multi-Label Classification Model,Triangle Visualization,Byte Tracker,Crop Visualization,Model Comparison Visualization,Detection Event Log,Stitch OCR Detections,OC-SORT Tracker,Overlap Filter,Track Class Lock,Heatmap Visualization,Webhook Sink,Florence-2 Model,Time in Zone,Detections Stabilizer,Dynamic Crop,Event Writer,Perspective Correction,Segment Anything 2 Model,Line Counter,GLM-OCR,Instance Segmentation Model,SAM2 Video Tracker,Multi-Label Classification Model,Circle Visualization,Line Counter,Distance Measurement,Ellipse Visualization,Time in Zone,Byte Tracker,Florence-2 Model,Pixelate Visualization,Detections Transformation,Roboflow Custom Metadata,Overlap Analysis,SORT Tracker,ByteTrack Tracker,Corner Visualization,Label Visualization,SmolVLM2,Path Deviation,Object Detection Model,Single-Label Classification Model,Qwen2.5-VL,Velocity,Per-Class Confidence Filter,Semantic Segmentation Model,Roboflow Vision Events,Trace Visualization,Color Visualization,Roboflow Dataset Upload,Detections Consensus,PTZ Tracking (ONVIF),Dot Visualization,Keypoint Detection Model,Detection Offset,SAM 3,Bounding Box Visualization,Camera Focus,Single-Label Classification Model,SAM3 Video Tracker,Moondream2,Detections Combine,Object Detection Model,Roboflow Dataset Upload,Detections Merge,Mask Area Measurement,Stitch OCR Detections,BoT-SORT Tracker,Detections Classes Replacement,Instance Segmentation Model,Model Monitoring Inference Aggregator,Byte Tracker,SAM 3,Multi-Label Classification Model,Detections Stitch,Semantic Segmentation Model,Size Measurement,Qwen3.5,Keypoint Detection Model,Icon Visualization,Blur Visualization,Time in Zone,Qwen3.5-VL,GeoTag Detection,Single-Label Classification Model,Instance Segmentation Model,Object Detection Model,Background Color Visualization,Detections List Roll-Up
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
Object Detection Model in version v3 has.
Bindings
-
input
images(image): The image to infer on..model_id(roboflow_model_id): Roboflow model identifier..confidence_mode(string): How confidence thresholds are determined..custom_confidence(float_zero_to_one): Custom confidence threshold for predictions..class_filter(list_of_values): List of accepted classes. Classes must exist in the model's training set..iou_threshold(float_zero_to_one): Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more..max_detections(integer): Maximum number of detections to return..class_agnostic_nms(boolean): Boolean flag to specify if NMS is to be used in class-agnostic mode..max_candidates(integer): Maximum number of candidates as NMS input to be taken into account..disable_active_learning(boolean): Boolean flag to disable project-level active learning for this block..active_learning_target_dataset(roboflow_project): Target dataset for active learning, if enabled..
-
output
inference_id(inference_id): Inference identifier.predictions(object_detection_prediction): Prediction with detected bounding boxes in form of sv.Detections(...) object.model_id(roboflow_model_id): Roboflow model id.
Example JSON definition of step Object Detection Model in version v3
{
"name": "<your_step_name_here>",
"type": "roboflow_core/roboflow_object_detection_model@v3",
"images": "$inputs.image",
"model_id": "my_project/3",
"confidence_mode": "<block_does_not_provide_example>",
"custom_confidence": 0.3,
"class_filter": [
"a",
"b",
"c"
],
"iou_threshold": 0.4,
"max_detections": 300,
"class_agnostic_nms": true,
"max_candidates": 3000,
"disable_active_learning": true,
"active_learning_target_dataset": "my_project"
}
v2¶
Class: RoboflowObjectDetectionModelBlockV2 (there are multiple versions of this block)
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 inference on a object-detection model hosted on or uploaded to Roboflow.
You can query any model that is private to your account, or any public model available on Roboflow Universe.
You will need to set your Roboflow API key in your Inference environment to use this block. To learn more about setting your Roboflow API key, refer to the Inference documentation.
Type identifier¶
Use the following identifier in step "type" field: roboflow_core/roboflow_object_detection_model@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 |
Roboflow model identifier.. | ✅ |
confidence |
float |
Confidence threshold for predictions.. | ✅ |
class_filter |
List[str] |
List of accepted classes. Classes must exist in the model's training set.. | ✅ |
iou_threshold |
float |
Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more.. | ✅ |
max_detections |
int |
Maximum number of detections to return.. | ✅ |
class_agnostic_nms |
bool |
Boolean flag to specify if NMS is to be used in class-agnostic mode.. | ✅ |
max_candidates |
int |
Maximum number of candidates as NMS input to be taken into account.. | ✅ |
disable_active_learning |
bool |
Boolean flag to disable project-level active learning for this block.. | ✅ |
active_learning_target_dataset |
str |
Target dataset for active learning, if enabled.. | ✅ |
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 Object Detection Model in version v2.
- inputs:
Absolute Static Crop,VLM As Classifier,Qwen-VL,Triangle Visualization,Model Comparison Visualization,Detection Event Log,Llama 3.2 Vision,Heatmap Visualization,Stitch Images,Keypoint Visualization,Anthropic Claude,Webhook Sink,Google Gemini,QR Code Generator,Instance Segmentation Model,VLM As Detector,PLC ModbusTCP,Twilio SMS/MMS Notification,Line Counter,Distance Measurement,Ellipse Visualization,Single-Label Classification Model,Keypoint Detection Model,Pixelate Visualization,Corner Visualization,Buffer,Template Matching,Llama 3.2 Vision,JSON Parser,Qwen 3.5 API,Roboflow Visual Search,Semantic Segmentation Model,Morphological Transformation,Color Visualization,Roboflow Dataset Upload,Dimension Collapse,Morphological Transformation,Camera Calibration,Motion Detection,Clip Comparison,Bounding Box Visualization,OpenAI,Halo Visualization,Object Detection Model,Identify Changes,Contrast Equalization,Instance Segmentation Model,Model Monitoring Inference Aggregator,Multi-Label Classification Model,MQTT Writer,PLC Reader,Image Threshold,Blur Visualization,Line Counter Visualization,Google Gemini,Background Color Visualization,Microsoft SQL Server Sink,Background Subtraction,VLM As Classifier,Identify Outliers,Object Detection Model,Twilio SMS Notification,OPC UA Writer Sink,Polygon Zone Visualization,Google Gemma API,Keypoint Detection Model,Reference Path Visualization,Image Stack,Stability AI Inpainting,Multi-Label Classification Model,Crop Visualization,Anthropic Claude,Image Convert Grayscale,Dynamic Zone,Camera Focus,Image Slicer,Florence-2 Model,Polygon Visualization,MoonshotAI Kimi,Dynamic Crop,Qwen 3.6 API,Event Writer,Perspective Correction,Line Counter,Polygon Visualization,Clip Comparison,Roboflow Visual Search Classifier,Circle Visualization,Slack Notification,Contrast Enhancement,PLC EthernetIP,Grid Visualization,Local File Sink,Florence-2 Model,Roboflow Custom Metadata,Image Blur,Label Visualization,Anthropic Claude,Pixel Color Count,Roboflow Vision Events,Image Preprocessing,Trace Visualization,Email Notification,SIFT Comparison,Stability AI Outpainting,Detections Consensus,PTZ Tracking (ONVIF),Dot Visualization,Roboflow Asset Library Attributes,Halo Visualization,Classification Label Visualization,Mask Visualization,Image Contours,Camera Focus,Single-Label Classification Model,OpenRouter,S3 Sink,Stability AI Image Generation,VLM As Detector,Google Gemma,Relative Static Crop,Roboflow Dataset Upload,OpenAI,Image Slicer,Google Gemini,Email Notification,Text Display,Semantic Segmentation Model,Size Measurement,Depth Estimation,Icon Visualization,OpenAI,GeoTag Detection,SIFT Comparison,SIFT,Instance Segmentation Model,PLC Writer,Detections List Roll-Up,MoonshotAI Kimi - outputs:
Path Deviation,Detections Filter,SAM 3,Instance Segmentation Model,Keypoint Detection Model,Qwen-VL,SAM 3 Interactive,Qwen3-VL,Multi-Label Classification Model,Triangle Visualization,Byte Tracker,Crop Visualization,Model Comparison Visualization,Detection Event Log,Stitch OCR Detections,OC-SORT Tracker,Overlap Filter,Track Class Lock,Heatmap Visualization,Webhook Sink,Florence-2 Model,Time in Zone,Detections Stabilizer,Dynamic Crop,Event Writer,Perspective Correction,Segment Anything 2 Model,Line Counter,GLM-OCR,Instance Segmentation Model,SAM2 Video Tracker,Multi-Label Classification Model,Circle Visualization,Line Counter,Distance Measurement,Ellipse Visualization,Time in Zone,Byte Tracker,Florence-2 Model,Pixelate Visualization,Detections Transformation,Roboflow Custom Metadata,Overlap Analysis,SORT Tracker,ByteTrack Tracker,Corner Visualization,Label Visualization,SmolVLM2,Path Deviation,Object Detection Model,Single-Label Classification Model,Qwen2.5-VL,Velocity,Per-Class Confidence Filter,Semantic Segmentation Model,Roboflow Vision Events,Trace Visualization,Color Visualization,Roboflow Dataset Upload,Detections Consensus,PTZ Tracking (ONVIF),Dot Visualization,Keypoint Detection Model,Detection Offset,SAM 3,Bounding Box Visualization,Camera Focus,Single-Label Classification Model,SAM3 Video Tracker,Moondream2,Detections Combine,Object Detection Model,Roboflow Dataset Upload,Detections Merge,Mask Area Measurement,Stitch OCR Detections,BoT-SORT Tracker,Detections Classes Replacement,Instance Segmentation Model,Model Monitoring Inference Aggregator,Byte Tracker,SAM 3,Multi-Label Classification Model,Detections Stitch,Semantic Segmentation Model,Size Measurement,Qwen3.5,Keypoint Detection Model,Icon Visualization,Blur Visualization,Time in Zone,Qwen3.5-VL,GeoTag Detection,Single-Label Classification Model,Instance Segmentation Model,Object Detection Model,Background Color Visualization,Detections List Roll-Up
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
Object Detection Model in version v2 has.
Bindings
-
input
images(image): The image to infer on..model_id(roboflow_model_id): Roboflow model identifier..confidence(float_zero_to_one): Confidence threshold for predictions..class_filter(list_of_values): List of accepted classes. Classes must exist in the model's training set..iou_threshold(float_zero_to_one): Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more..max_detections(integer): Maximum number of detections to return..class_agnostic_nms(boolean): Boolean flag to specify if NMS is to be used in class-agnostic mode..max_candidates(integer): Maximum number of candidates as NMS input to be taken into account..disable_active_learning(boolean): Boolean flag to disable project-level active learning for this block..active_learning_target_dataset(roboflow_project): Target dataset for active learning, if enabled..
-
output
inference_id(inference_id): Inference identifier.predictions(object_detection_prediction): Prediction with detected bounding boxes in form of sv.Detections(...) object.model_id(roboflow_model_id): Roboflow model id.
Example JSON definition of step Object Detection Model in version v2
{
"name": "<your_step_name_here>",
"type": "roboflow_core/roboflow_object_detection_model@v2",
"images": "$inputs.image",
"model_id": "my_project/3",
"confidence": 0.3,
"class_filter": [
"a",
"b",
"c"
],
"iou_threshold": 0.4,
"max_detections": 300,
"class_agnostic_nms": true,
"max_candidates": 3000,
"disable_active_learning": true,
"active_learning_target_dataset": "my_project"
}
v1¶
Class: RoboflowObjectDetectionModelBlockV1 (there are multiple versions of this block)
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 inference on a object-detection model hosted on or uploaded to Roboflow.
You can query any model that is private to your account, or any public model available on Roboflow Universe.
You will need to set your Roboflow API key in your Inference environment to use this block. To learn more about setting your Roboflow API key, refer to the Inference documentation.
Type identifier¶
Use the following identifier in step "type" field: roboflow_core/roboflow_object_detection_model@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 |
Roboflow model identifier.. | ✅ |
confidence |
float |
Confidence threshold for predictions.. | ✅ |
class_filter |
List[str] |
List of accepted classes. Classes must exist in the model's training set.. | ✅ |
iou_threshold |
float |
Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more.. | ✅ |
max_detections |
int |
Maximum number of detections to return.. | ✅ |
class_agnostic_nms |
bool |
Boolean flag to specify if NMS is to be used in class-agnostic mode.. | ✅ |
max_candidates |
int |
Maximum number of candidates as NMS input to be taken into account.. | ✅ |
disable_active_learning |
bool |
Boolean flag to disable project-level active learning for this block.. | ✅ |
active_learning_target_dataset |
str |
Target dataset for active learning, if enabled.. | ✅ |
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 Object Detection Model in version v1.
- inputs:
Absolute Static Crop,VLM As Classifier,Qwen-VL,Triangle Visualization,Model Comparison Visualization,Detection Event Log,Llama 3.2 Vision,Heatmap Visualization,Stitch Images,Keypoint Visualization,Anthropic Claude,Webhook Sink,Google Gemini,QR Code Generator,Instance Segmentation Model,VLM As Detector,PLC ModbusTCP,Twilio SMS/MMS Notification,Line Counter,Distance Measurement,Ellipse Visualization,Single-Label Classification Model,Keypoint Detection Model,Pixelate Visualization,Corner Visualization,Buffer,Template Matching,Llama 3.2 Vision,JSON Parser,Qwen 3.5 API,Roboflow Visual Search,Semantic Segmentation Model,Morphological Transformation,Color Visualization,Roboflow Dataset Upload,Dimension Collapse,Morphological Transformation,Camera Calibration,Motion Detection,Clip Comparison,Bounding Box Visualization,OpenAI,Halo Visualization,Object Detection Model,Identify Changes,Contrast Equalization,Instance Segmentation Model,Model Monitoring Inference Aggregator,Multi-Label Classification Model,MQTT Writer,PLC Reader,Image Threshold,Blur Visualization,Line Counter Visualization,Google Gemini,Background Color Visualization,Microsoft SQL Server Sink,Background Subtraction,VLM As Classifier,Identify Outliers,Object Detection Model,Twilio SMS Notification,OPC UA Writer Sink,Polygon Zone Visualization,Google Gemma API,Keypoint Detection Model,Reference Path Visualization,Image Stack,Stability AI Inpainting,Multi-Label Classification Model,Crop Visualization,Anthropic Claude,Image Convert Grayscale,Dynamic Zone,Camera Focus,Image Slicer,Florence-2 Model,Polygon Visualization,MoonshotAI Kimi,Dynamic Crop,Qwen 3.6 API,Event Writer,Perspective Correction,Line Counter,Polygon Visualization,Clip Comparison,Roboflow Visual Search Classifier,Circle Visualization,Slack Notification,Contrast Enhancement,PLC EthernetIP,Grid Visualization,Local File Sink,Florence-2 Model,Roboflow Custom Metadata,Image Blur,Label Visualization,Anthropic Claude,Pixel Color Count,Roboflow Vision Events,Image Preprocessing,Trace Visualization,Email Notification,SIFT Comparison,Stability AI Outpainting,Detections Consensus,PTZ Tracking (ONVIF),Dot Visualization,Roboflow Asset Library Attributes,Halo Visualization,Classification Label Visualization,Mask Visualization,Image Contours,Camera Focus,Single-Label Classification Model,OpenRouter,S3 Sink,Stability AI Image Generation,VLM As Detector,Google Gemma,Relative Static Crop,Roboflow Dataset Upload,OpenAI,Image Slicer,Google Gemini,Email Notification,Text Display,Semantic Segmentation Model,Size Measurement,Depth Estimation,Icon Visualization,OpenAI,GeoTag Detection,SIFT Comparison,SIFT,Instance Segmentation Model,PLC Writer,Detections List Roll-Up,MoonshotAI Kimi - outputs:
Detections Filter,Qwen-VL,Cache Get,Triangle Visualization,Byte Tracker,Model Comparison Visualization,Detection Event Log,Stitch OCR Detections,OC-SORT Tracker,YOLO-World Model,Track Class Lock,Llama 3.2 Vision,Anthropic Claude,Keypoint Visualization,Heatmap Visualization,Webhook Sink,Detections Stabilizer,Google Gemini,GLM-OCR,QR Code Generator,Instance Segmentation Model,MoonshotAI Kimi,SAM2 Video Tracker,Twilio SMS/MMS Notification,Line Counter,Distance Measurement,Ellipse Visualization,Single-Label Classification Model,Pixelate Visualization,Detections Transformation,Corner Visualization,Overlap Analysis,Llama 3.2 Vision,Qwen 3.5 API,Velocity,LMM,Per-Class Confidence Filter,Roboflow Visual Search,OpenAI-Compatible LLM,Morphological Transformation,Color Visualization,Roboflow Dataset Upload,Morphological Transformation,Detection Offset,Google Vision OCR,Clip Comparison,Bounding Box Visualization,OpenAI,Halo Visualization,SAM3 Video Tracker,Detections Combine,Contrast Equalization,Mask Area Measurement,Detections Classes Replacement,Instance Segmentation Model,Model Monitoring Inference Aggregator,MQTT Writer,Multi-Label Classification Model,Cache Set,Image Threshold,Blur Visualization,Line Counter Visualization,Google Gemini,Microsoft SQL Server Sink,Object Detection Model,Path Deviation,Twilio SMS Notification,OPC UA Writer Sink,SAM 3,Instance Segmentation Model,Google Gemma API,Polygon Zone Visualization,Keypoint Detection Model,Reference Path Visualization,SAM 3 Interactive,Stability AI Inpainting,Perception Encoder Embedding Model,Crop Visualization,Anthropic Claude,Overlap Filter,Florence-2 Model,Polygon Visualization,Time in Zone,MoonshotAI Kimi,Dynamic Crop,Qwen 3.6 API,Event Writer,Perspective Correction,Segment Anything 2 Model,Line Counter,Polygon Visualization,Roboflow Visual Search Classifier,Circle Visualization,Slack Notification,CogVLM,Time in Zone,Byte Tracker,LMM For Classification,Qwen3.5-VL,Florence-2 Model,OpenAI,Roboflow Custom Metadata,Local File Sink,SORT Tracker,Label Visualization,Image Blur,ByteTrack Tracker,Seg Preview,Path Deviation,Anthropic Claude,Pixel Color Count,Roboflow Vision Events,Image Preprocessing,Trace Visualization,Email Notification,SIFT Comparison,Stability AI Outpainting,Detections Consensus,PTZ Tracking (ONVIF),Dot Visualization,Current Time,Roboflow Asset Library Attributes,Halo Visualization,SAM 3,Classification Label Visualization,Camera Focus,OpenRouter,Moondream2,S3 Sink,Stability AI Image Generation,Google Gemma,Roboflow Dataset Upload,CLIP Embedding Model,OpenAI,Detections Merge,Stitch OCR Detections,Google Gemini,BoT-SORT Tracker,Email Notification,Byte Tracker,SAM 3,Detections Stitch,Text Display,Semantic Segmentation Model,Size Measurement,Depth Estimation,Icon Visualization,OpenAI,Time in Zone,GeoTag Detection,Instance Segmentation Model,Background Color Visualization,Detections List Roll-Up,Mask Visualization
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
Object Detection Model in version v1 has.
Bindings
-
input
images(image): The image to infer on..model_id(roboflow_model_id): Roboflow model identifier..confidence(float_zero_to_one): Confidence threshold for predictions..class_filter(list_of_values): List of accepted classes. Classes must exist in the model's training set..iou_threshold(float_zero_to_one): Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more..max_detections(integer): Maximum number of detections to return..class_agnostic_nms(boolean): Boolean flag to specify if NMS is to be used in class-agnostic mode..max_candidates(integer): Maximum number of candidates as NMS input to be taken into account..disable_active_learning(boolean): Boolean flag to disable project-level active learning for this block..active_learning_target_dataset(roboflow_project): Target dataset for active learning, if enabled..
-
output
inference_id(string): String value.predictions(object_detection_prediction): Prediction with detected bounding boxes in form of sv.Detections(...) object.
Example JSON definition of step Object Detection Model in version v1
{
"name": "<your_step_name_here>",
"type": "roboflow_core/roboflow_object_detection_model@v1",
"images": "$inputs.image",
"model_id": "my_project/3",
"confidence": 0.3,
"class_filter": [
"a",
"b",
"c"
],
"iou_threshold": 0.4,
"max_detections": 300,
"class_agnostic_nms": true,
"max_candidates": 3000,
"disable_active_learning": true,
"active_learning_target_dataset": "my_project"
}