Build Complex Vision Applications with Workflows

Prototype, experiment, test, integrate, and deploy pipelines to production
Create a Workflow
cogs
Integrated workflow builder and deployment infrastructure ensures that what you configure can be deployed
cctv
Deploy using a fully managed infrastructure with an API endpoint or on-device, internet connection optional, without the headache of environment management
cubes
Combine custom models, open source models, LLM APIs, pre-built logic, and external applications

How to Build a Workflow

Learn how to use a low-code open source platform to simplify building and deploying vision AI applications.
1
cube icon

Choose a Block

Choose from 40+ pre-built blocks that let you use custom models, open source models, LLM APIs, pre-built logic, and external applications. Blocks can be models from OpenAI or Meta AI, applications like Google Sheets or Pager Duty, and logic like filtering or cropping.
2
connected blocks icon

Connect Blocks

Each block can receive inputs, execute code, and send outputs to the next block in your Workflow. You can use the drag-and-drop UI to configure connections and see the JSON definitions of what’s happening behind the scenes.
3
rocket icon

Deploy Workflows

You’ll receive an output of the final result from your Workflow and the format you want it delivered in, like JSON. Once your Workflow produces sufficient results, you can use the Workflow as a hosted API endpoint or self-host in your own cloud, on-prem, or at the edge.

Workflow Templates

Explore pre-made templates for specific use cases.
  • Image Slicer
  • Roboflow Object Detection Model
Detect people in a large image
Add to workspace
  • Image Slicer
  • Roboflow Object Detection Model
Detect vehicles in a large image
Add to workspace
  • Roboflow Object Detection Model
  • Super Annotator
Detect people
Add to workspace
  • Roboflow Object Detection Model
  • Super Annotator
Detect vehicles in an image or video
Add to workspace
  • Roboflow Object Detection Model
  • Super Annotator
Detect fish
Add to workspace
  • Roboflow Object Detection Model
  • Super Annotator
Detect potholes
Add to workspace
  • Yolo World Model
  • Super Annotator
Detect common animals
Add to workspace
  • Roboflow Object Detection Model
  • Super Annotator
Remove the background from an image
Add to workspace
View All Workflow Templates

Explore Popular Combinations

Save time with powerful model and logic connections to solve complex problems.
Model
Sink

YOLOv7 Instance Segmentation to Active Learning Data Collector

Build a computer vision workflow that connects YOLOv7 Instance Segmentation to Active Learning Data Collector.
Model
Sink

YOLOv5 to Active Learning Data Collector

Build a computer vision workflow that connects YOLOv5 to Active Learning Data Collector.
Model
Sink

YOLOv8 to Active Learning Data Collector

Build a computer vision workflow that connects YOLOv8 to Active Learning Data Collector.
Model
Sink

YOLOv8 Keypoint Detection to Active Learning Data Collector

Build a computer vision workflow that connects YOLOv8 Keypoint Detection to Active Learning Data Collector.
Model
Sink

YOLOv8 Instance Segmentation to Active Learning Data Collector

Build a computer vision workflow that connects YOLOv8 Instance Segmentation to Active Learning Data Collector.
Model
Sink

YOLO-NAS to Active Learning Data Collector

Build a computer vision workflow that connects YOLO-NAS to Active Learning Data Collector.

Customize Your Pipeline

Connect models from OpenAI or Meta AI, applications like Slack or Pager Duty, and logic like filtering or cropping.
View All Blocks

Simplify Building and Deploying Vision AI Applications

Integrated workflow builder and deployment ensures what you configure runs in production
Get started for free
list of model blocks

Models

Integrated workflow builder and deployment ensures what is configured runs in production
list of transformation blocks

Transformations

Dynamically process inputs to improve the results of your model predictions
list of branching blocks

Branching

Add custom or pre-built logic for flow-control and processing
list of customization blocks

Customization

Create and add custom blocks to your pipeline
list of integration blocks

Integrations

Connect your pipeline to external applications

Deploy Workflows at Scale

Roboflow powers millions of daily inferences for the world’s largest enterprises on-device and in the cloud
Deploy your Workflows directly on fully managed infrastructure through an infinitely-scalable API endpoint for high volume workloads
Run Workflows on-device, internet connection optional, without the headache of environment management, dependencies, and managing CUDA versions.
Isolate dependencies in your software by using the Python SDK or HTTP API to operate and maintain your Workflows separate from other logic within your codebase
Supported devices include ARM CPU, x86 CPU, NVIDIA GPU, and NVIDIA Jetson

Building Blocks for Custom Pipelines

Blocks receive inputs, execute code, and send outputs to the next block in your Workflow
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
icon modelTransformation

AWS Secrets Retriever

Retrieve and manage secrets from AWS Secrets Manager for use in workflow automation.
Transformation
icon modelTransformation

Absolute Static Crop

Use absolute coordinates for cropping.
Transformation
icon modelSink

Active Learning Data Collector

Collect data and predictions that flow through workflows for use in active learning.
Sink
icon modelNotifications

Azure Sink

This block sends email alerts with images and predictions.
Notifications
icon model

Background Color Visualization

Paints a mask over all areas outside of detected regions in an image.
icon model

Barcode Detection

Detect and read barcodes in an image.
icon modelModel

Barcode Detector

Run Optical Character Recognition on a model.
Model
icon model

Blur Visualization

Blurs detected objects in an image.
icon model

Bounding Box Visualization

Draws a box around detected objects in an image.
icon modelTracker

Byte Tracker

Track and update object positions across video frames using ByteTrack.
Tracker
icon modelSink

CSV Sink

Export prediction data to a CSV file for various prediction kinds, including object detection and instance segmentation.
Sink
icon model

Circle Visualization

Draws a circle around detected objects in an image.
icon modelModel

Clip Comparison

Compare CLIP image and text embeddings.
Model
icon modelModel

CogVLM

CogVLM is a type of LMM.
Model
icon model

Color Visualization

Paints a solid color on detected objects in an image.
icon modelFlow Control

Condition

Control the flow of a workflow based on the result of a step.
Flow Control
icon model

Continue If

Continue a branch of logic if a condition is met.
icon model

Corner Visualization

Draws the corners of detected objects in an image.
icon modelTransformation

Crop

Create dynamic crops from a detections model.
Transformation
icon model

Crop Visualization

Draws scaled up crops of detections on the scene.
icon modelTransformation

Detection Filter

Filter predictions from detection models based on defined conditions.
Transformation
icon modelTransformation

Detection Offset

Apply a fixed offset on the width and height of detections.
Transformation
icon model

Detections Classes Replacement

Replaces classes of detections with classes predicted by a chained classification model
icon modelFusion

Detections Consensus

Combine predictions from multiple detections models to make a decision about object presence.
Fusion
icon model

Detections Filter

Conditionally filter out model predictions.
icon model

Detections Stitch

Merges detections made against multiple pieces of input image into single detection.
icon model

Detections Transformation

Apply transformations on detected bounding boxes.
icon model

DetectionsConsensus

Combine predictions from multiple detections models to make a decision about object presence.
icon model

Dimension Collapse

Collapses dimensionality level by aggregation of nested data into list
icon model

Dominant Color

Get the dominant color of an image in RGB format.
icon model

Dot Visualization

Draws dots on an image at specific coordinates based on provided detections.
icon model

Dynamic Crop

Crop an image using bounding boxes from a detection model.
icon model

Dynamic Zone

Simplify polygons so they are geometrically convex and simplify them to contain only requested amount of vertices
icon model

Ellipse Visualization

Draws ellipses that highlight detected objects in an image.
icon model

Expression

Evaluate a condition and return a response.
icon modelTransformation

Filter

Conditionally filter and output data fields based on boolean conditions from a preceding workflow step.
Transformation
icon model

First Non Empty Or Default

Takes first non-empty data element or default
icon modelModel

GPT-4 with Vision

GPT-4 with Vision is a type of LMM.
Model
icon model

Halo Visualization

Paints a halo around detected objects in an image.
icon model

Image Blur

Apply a blur to an image.
icon model

Image Contours

Find and count the contours on an image.
icon model

Image Convert Grayscale

Convert an RGB image to grayscale.
icon model

Image Slicer

Use SAHI to run sliced inference with a computer vision model.
icon model

Image Threshold

Apply a threshold to an image.
icon model

Instance Segmentation Model

Predict the shape, size, and location of objects.
icon modelTransformation

Kafka Publish

Publish data to a Kafka topic, with server connection and topic configuration, ensuring data propagation within distributed systems.
Transformation
icon model

Keypoint Detection Model

Predict skeletons on objects.
icon modelModel

LMM

Run a large language model.
Model
icon modelModel

LMM For Classification

Run a large language model for classification.
Model
icon model

Label Visualization

Draws labels on an image at specific coordinates based on provided detections.
icon modelSink

MQTT Publisher

Publish serialized data payloads to an MQTT broker, handling connection management and error reporting.
Sink
icon model

Mask Visualization

Paints a mask over detected objects in an image.
icon model

Multi-Label Classification Model

Apply multiple tags to an image.
icon modelTransformation

Non Empty Predictions Tagger

Tag batches of images with boolean flags indicating the presence or absence of predictions for various detection types.
Transformation
icon modelModel

OCR Model

Run Optical Character Recognition on a model.
Model
icon model

Object Detection Model

Predict the location of objects with bounding boxes.
icon model

OpenAI

Run OpenAI's GPT-4 with Vision
icon modelNotifications

Pager Duty

Trigger alerts on PagerDuty based on configurable conditions, sending notifications with customized details about system events.
Notifications
icon model

Perspective Correction

Correct coordinates of detections from plane defined by given polygon to straight rectangular plane of given width and height
icon model

Pixel Color Count

Count the number of pixels that match a specific color within a given tolerance.
icon model

Pixelate Visualization

Pixelates detected objects in an image.
icon model

Polygon Visualization

Draws a polygon around detected objects in an image.
icon modelTransformation

Prediction Alarm

Monitor and trigger an alarm if the count of specified prediction classes within a defined timeframe exceeds a set threshold.
Transformation
icon model

Property Definition

Define a variable from model predictions, such as the class names, confidences, or number of detections.
icon model

QR Code Detection

Detect and read QR codes in an image.
icon modelModel

QR Code Detector

Detect the location of QR codes in an image.
Model
icon modelTransformation

Random Sampler

Randomly sample images based on a specified percentage, triggering actions for selected images within workflow processes.
Transformation
icon modelTransformation

Relative Static Crop

Use relative coordinates for cropping.
Transformation
icon modelModel

Roboflow Classification Model

Run a classification model.
Model
icon model

Roboflow Custom Metadata

Add custom metadata to Roboflow Model Monitoring dashboard
icon model

Roboflow Dataset Upload

Save images and predictions in your Roboflow Dataset
icon modelModel

Roboflow Instance Segmentation Model

Run an instance segmentation model.
Model
icon modelModel

Roboflow Keypoint Detection Model

Run inference on a keypoint detection model.
Model
icon modelModel

Roboflow Multi Label Classification Model

Run a multi-label classification model.
Model
icon modelModel

Roboflow Object Detection Model

Detect objects using an object detection model.
Model
icon modelUtility

S3 Upload

Send images and prediction files to AWS S3.
Utility
icon model

SIFT

Apply SIFT to an image.
icon model

SIFT Comparison

Compare SIFT descriptors from multiple images.
icon modelNotifications

SMS

Send SMS notifications through Twilio based on configurable conditions, utilizing dynamic message content and sender/receiver details.
Notifications
icon model

Segment Anything 2

Use SAM 2 to segment objects in an image.
icon model

Segment Anything 2 Model

Convert bounding boxes to polygons, or run SAM2 on an entire image to generate a mask.
icon modelSink

Serializer

Convert object detection predictions into serialized format, facilitating data transfer or storage within workflow pipelines.
Sink
icon model

Single-Label Classification Model

Apply a single tag to an image.
icon modelNotifications

Slack

Send notifications to a specified Slack channel based on conditions, using customizable message content and Slack API tokens.
Notifications
icon modelSink

Super Annotator

Enhance images with visual annotations based on detection predictions, combining bounding boxes and labels for clarity.
Sink
icon model

Template Matching

Looks for instances of template in specific image
icon model

Triangle Visualization

Draws triangle markers on an image at specific coordinates based on provided detections.
icon modelSink

Webhook Publish

Securely transmit data to specified webhooks using HMAC-SHA512 for payload authentication, ensuring data integrity and confidentiality.
Sink
icon modelModel

YOLO-NAS

YOLO-NAS is a type of Roboflow Object Detection Model.
Model
icon model

YOLO-World Model

Run a zero-shot object detection model.
icon modelModel

YOLOv5

YOLOv5 is a type of Roboflow Object Detection Model.
Model
icon modelModel

YOLOv5 Instance Segmentation

YOLOv5 Instance Segmentation is a type of Roboflow Instance Segmentation Model.
Model
icon modelModel

YOLOv7 Instance Segmentation

YOLOv7 Instance Segmentation is a type of Roboflow Instance Segmentation Model.
Model
icon modelModel

YOLOv8

YOLOv8 is a type of Roboflow Object Detection Model.
Model
icon modelModel

YOLOv8 Classification

YOLOv8 Classification is a type of Roboflow Classification Model.
Model
icon modelModel

YOLOv8 Instance Segmentation

YOLOv8 Instance Segmentation is a type of Roboflow Instance Segmentation Model.
Model
icon modelModel

YOLOv8 Keypoint Detection

YOLOv8 Keypoint Detection is a type of Roboflow Keypoint Detection Model.
Model
icon modelModel

YOLOv9

YOLOv9 is a type of Roboflow Object Detection Model.
Model
icon modelModel

Yolo World Model

Run a zero-shot object detection model.
Model