AI Hub + PySDK Workflow
Use AI Hub to explore hardware, browse public models, and run hosted inference. Then use the same models in your PySDK application when you are ready to build.
Explore integration >
AI Hub for Edge AI Development
Transform your operations with scalable, cost-efficient AI that runs in the cloud or at the edge. Automate workflows, gain real-time insights, and accelerate innovation across the entire ecosystem.
99.9%
Uptime SLA
100+
Supported Models
10+
Hardware Runtimes
Browser Inference
PySDK Quickstart
Load a model from a zoo, connect to an inference target, and run your first inference with structured results.
Read the quickstart >PySDK Installation
Install the package and set up optional components for your runtime and deployment target.
Install PySDK >Run Models Directly in the Browser
Enable interactive AI experiences directly in modern browsers without requiring a backend service for every request. This makes it easier to build demos, client-side tools, and privacy-friendly applications with faster user feedback.
Learn more >Public Model Zoos
Face Detection
Detect and localize faces in images and video streams for analytics, tracking, and vision workflows.
Segmentation
Separate objects or regions at pixel level for scene understanding, masking, and precise analysis.
PPE Detection
Identify safety equipment such as helmets, vests, and gloves in industrial and workplace environments.
Image Classification
Assign labels to images for recognition, sorting, and visual categorization tasks.
Browse production-ready AI models
Explore curated public model zoos with support for common computer vision tasks. Quickly evaluate models, compare outputs, and move from discovery to integration using the same workflow.
Get StartedObject Detection
Find and classify objects within frames for retail, logistics, surveillance, and automation use cases.
License Plate Detection
Locate license plates in traffic and parking scenarios for vehicle monitoring and access workflows.
Pose Estimation
Track body keypoints and movement for fitness, ergonomics, behavior analysis, and interaction systems.
OCR
Extract printed or structured text from images and documents for search, validation, and automation.
Cloud Compiler
Model Optimization
Compile and optimize models in the cloud for supported runtimes, reducing manual setup and accelerating deployment workflows.
Hardware-Aware Compilation
Prepare models for different accelerator targets using a unified cloud-based pipeline that simplifies multi-platform execution.
Scalable Cloud Pipeline
Run compilation workflows in the cloud to support large model libraries, continuous updates, and faster iteration across teams.
Cross-Platform Deployment
Use compiled models across edge devices, browser workflows, and server environments with consistent integration paths.
Start building with DeGirum today
Access SDKs, model zoo, and all Application Packages — from cloud or edge.