--- hide: - toc ---

Workflow Blocks

Models

Object Detection Model
Predict the location of objects with bounding boxes.
Instance Segmentation Model
Predict the shape, size, and location of objects.
Single-Label Classification Model
Apply a single tag to an image.
Multi-Label Classification Model
Apply multiple tags to an image.
Semantic Segmentation Model
Assign a class label to every pixel in the image.
Keypoint Detection Model
Predict skeletons on objects.
Anthropic Claude
Run Anthropic Claude model with vision capabilities.
Google Gemini
Run Google's Gemini model with vision capabilities.
OpenAI
Run OpenAI's GPT models with vision capabilities.
Depth Estimation
Run Depth Estimation on an image.
Florence-2 Model
Run Florence-2 on an image
Qwen-VL
Run any Qwen vision model — natively or via OpenRouter.
SmolVLM2
Run SmolVLM2 on an image.
Moondream2
Run Moondream2 on an image.
OpenRouter
Run any OpenRouter model by pasting its model slug.
Qwen3.5
Run Qwen3.5 on an image.
YOLO-World Model
Run a zero-shot object detection model.
CogVLM
DEPRECATED! Run a self-hosted vision language model.
SAM 3
Run SAM3 with text prompts for zero-shot segmentation.
Seg Preview
Seg Preview
Segment Anything 2 Model
Convert bounding boxes to polygons, or run SAM2 on an entire image to generate a mask.
CLIP Embedding Model
Generate an embedding of an image or string.
Perception Encoder Embedding Model
Generate an embedding of an image or string.
Clip Comparison
Compare CLIP image and text embeddings.
OCR Model
Extract text from an image using DocTR optical character recognition.
GLM-OCR
Run GLM-OCR on an image to recognize text.
EasyOCR
Extract text from an image using EasyOCR optical character recognition.
Barcode Detection
Detect and read barcodes in an image.
QR Code Detection
Detect and read QR codes in an image.
Stability AI Inpainting
Use segmentation masks to inpaint objects within an image.
Stability AI Outpainting
Use object detection bounding box to crop the image and to outpaint within given directions.
Google Vision OCR
Detect text in images using Google Vision API
Stability AI Image Generation
generate new images from text, or create variations of existing images.
Llama 3.2 Vision
Run Llama 3.2 Vision via OpenRouter.
Google Gemma
Run Google's Gemma model with vision capabilities via OpenRouter.
OpenAI-Compatible LLM
Send prompts to any OpenAI-compatible API endpoint.
MoonshotAI Kimi
Run Moonshot AI Kimi vision-language models via OpenRouter.

Visualizations

Bounding Box Visualization
Draw a box around detected objects in an image.
Dot Visualization
Draw dots on an image at specific coordinates based on provided detections.
Polygon Visualization
Draw a polygon around detected objects in an image.
Label Visualization
Draw labels on an image at specific coordinates based on provided detections.
Classification Label Visualization
Visualize both single-label and multi-label classification predictions with customizable display options.
Background Color Visualization
Apply a mask to cover all areas outside the detected regions in an image.
Text Display
Display customizable text on an image with styling and positioning options.
Blur Visualization
Blur detected objects in an image.
Heatmap Visualization
Draw a heatmap based on detections in an image.
Circle Visualization
Draw a circle around detected objects in an image.
Icon Visualization
Draw icons on an image either at specific static coordinates or dynamically based on detections.
Color Visualization
Paint a solid color on detected objects in an image.
Corner Visualization
Draw the corners of detected objects in an image.
Crop Visualization
Draw scaled up crops of detections on the scene.
Ellipse Visualization
Draw ellipses that highlight detected objects in an image.
Halo Visualization
Paint a halo around detected objects in an image.
Mask Visualization
Apply a mask over detected objects in an image.
Pixelate Visualization
Pixelate detected objects in an image.
Triangle Visualization
Draw triangle markers on an image at specific coordinates based on provided detections.
Line Counter Visualization
Apply a mask over a line zone in an image.
Polygon Zone Visualization
Apply a mask over a polygon zone in an image.
Model Comparison Visualization
Visualize the difference between two models' detections.
Trace Visualization
Draw traces based on detections tracking results.
Reference Path Visualization
Draw a reference path in the image.
Keypoint Visualization
Draw keypoints on detected objects in an image.
Grid Visualization
Shows an array of images in a grid.

Logic and branching

Continue If
Conditionally stop execution of a branch.
Detections Filter
Conditionally filter out model predictions.
Overlap Filter
Filter objects overlapping some other class
Per-Class Confidence Filter
Filter detections by applying a per-class confidence threshold.
Inner Workflow
Run a nested workflow definition with parameters mapped from the parent workflow.
Rate Limiter
Limits the rate at which a branch of the Workflow will run.
Delta Filter
Allow the execution of workflow to proceed if the input value has changed.
Detections Consensus
Combine predictions from multiple detections models to make a decision about object presence.

Data storage

Roboflow Dataset Upload
Save images and predictions to your Roboflow Dataset.
Roboflow Asset Library Attributes
Update attributes and tags for Asset Library images.
Webhook Sink
Send a request to a remote API with Workflow results.
Roboflow Vision Events
Send vision events to the Roboflow Vision Events API.
CSV Formatter
Create CSV files with specified columns.
Local File Sink
Save data to a local file.
Microsoft SQL Server Sink
Save data to a Microsoft SQL Server database.
Data Aggregator
Aggregate workflow data to produce time-based statistics.
S3 Sink
Upload data to an AWS S3 bucket.

Notifications

Email Notification
Send notification via e-mail.
Twilio SMS/MMS Notification
Send SMS/MMS notifications via Twilio.
Slack Notification
Send notification via Slack.
Twilio SMS Notification
Send notification via Twilio SMS service.

Transformations

Dynamic Crop
Crop an image using bounding boxes from a detection model.
Absolute Static Crop
Crop an image using fixed pixel coordinates.
Camera Calibration
Remove camera lens distortions from an image using a calibration table.
QR Code Generator
Generate a QR code image from text input.
Relative Static Crop
Crop an image proportional (%) to its dimensions.
Detection Offset
Apply a padding around the width and height of detections.
Detections Transformation
Apply transformations on detected bounding boxes.
Bounding Rectangle
Find the minimal bounding box surrounding the detected polygon.
Detections Merge
Merge multiple detections into a single bounding box.
Detections Combine
Combines two sets of predictions into a single prediction.
Stitch Images
Stitch two images by common parts.

Classical computer vision

Image Preprocessing
Resize, flip, or rotate an image.
Template Matching
Locate instances of a given template within a specified image.
Dominant Color
Get the dominant color of an image in RGB format.
Pixel Color Count
Count the number of pixels that match a specific color within a given tolerance.
SIFT Comparison
Compare SIFT descriptors from multiple images.
Image Contours
Find and count the contours on an image.
SIFT
Apply SIFT to an image.
Contrast Equalization
Apply contrast equalization to an image.
Image Blur
Apply a blur to an image.
Morphological Transformation
Apply morphological transformation to color images.
Image Threshold
Apply a threshold to an image.
Image Convert Grayscale
Convert an RGB image to grayscale.
Camera Focus
Calculate a score to indicate how well-focused a camera is.
Size Measurement
Measure the dimensions of objects in relation to a reference object.
Contrast Enhancement
Enhance image contrast by normalizing the histogram to use the full dynamic range.
Distance Measurement
Calculate the distance between two bounding boxes on a 2D plane.
Mask Area Measurement
Measure the area of detected objects and optionally convert to real-world units.
Mask Edge Snap
Refine instance segmentation masks by snapping edges to detected boundaries.

Video

ByteTrack Tracker
Tracks objects across frames. Best for most scenes.
BoT-SORT Tracker
ByteTrack-style association with optional camera motion compensation (BoT-SORT).
SORT Tracker
Fast, lightweight object tracking. Works best when objects are clearly visible.
OC-SORT Tracker
Tracks objects through occlusion and unpredictable movement.
Time in Zone
Track object time in zone.
PTZ Tracking (ONVIF)
Control an ONVIF compatible PTZ camera to follow an object
Line Counter
Count detections passing a line.
Velocity
Calculate the velocity and speed of tracked objects with smoothing and unit conversion.
Path Deviation
Calculate Fréchet distance of object from the reference path.
Detections Stabilizer
Apply a smoothing algorithm to reduce noise and flickering across video frames.
Motion Detection
Detect motion in a video using OpenCV.
Background Subtraction
Subtract an image from its background history.
SAM2 Video Tracker
Segment and track objects across video frames with SAM2's streaming camera predictor.
Buffer
Returns an array of the last `length` values passed to it.
Image Stack
Accumulate compressed video frames into a fixed-size FIFO stack.
Identify Outliers
Identify outlier embeddings compared to prior data.
Identify Changes
Identify changes compared to prior data via embeddings.

Advanced

Property Definition
Define a variable from model predictions, such as the class names, confidences, or number of detections.
Expression
Create a specific output based on defined input variables and configured rules.
Perspective Correction
Adjust detection coordinates from a polygon-defined plane to a straight rectangular plane with specified width and height.
Stitch OCR Detections
Combines OCR detection results into a coherent text string by organizing detections spatially.
Dynamic Zone
Simplify polygons so they are geometrically convex and contain only the requested amount of vertices.
Cosine Similarity
Calculate the cosine similarity between two embeddings.
Detections Classes Replacement
Replace classes of detections with classes predicted by a chained classification model.
JSON Parser
Parse raw string into JSON.
VLM As Classifier
Parse a raw string into a classification prediction.
VLM As Detector
Parses raw string into object-detection prediction.
Dimension Collapse
Collapse dimensionality by aggregating nested data into a single list.
Detections List Roll-Up
Roll up multiple levels of dimensionality back to a single dimension.
First Non Empty Or Default
Take the first non-empty data element or the configured default value.
Roboflow Custom Metadata
Add custom metadata to the Roboflow Model Monitoring dashboard.
Model Monitoring Inference Aggregator
Periodically report an aggregated sample of inference results to Roboflow Model Monitoring.
Image Slicer
Tile the input image into a list of smaller images to perform small object detection.
Current Time
Output the current date and time for a given timezone.
Detections Stitch
Merges detections made against multiple pieces of input image into single detection.
Cache Get
Fetches a previously stored value from a cache entry.
Cache Set
Stores a value in a cache entry for later retrieval.
Environment Secrets Store
Fetch secrets from environmental variables.

Industrial

Event Writer
Write structured events to the Event Ingestion Service.
MQTT Writer
Publishes messages to an MQTT broker.
OPC UA Writer Sink
Writes data to an OPC UA server using the [asyncua](https://github.com/FreeOpcUa/opcua-asyncio) library for communication.
PLC EthernetIP
Generic PLC read/write block using pylogix over Ethernet/IP.
PLC ModbusTCP
Generic Modbus TCP read/write block using pymodbus.