Model Optimization
Optimize TensorFlow, PyTorch, and ONNX models for Deepgrid hardware with automated quantization and pruning.
Software
DGrid SDK
Comprehensive development toolkit for deploying AI models on Deepgrid semiconductor platforms. Optimize, deploy, and scale your edge AI applications with ease.
End-to-end integrated software stack
The DGrid SDK is purpose-built for Computer Vision at the Edge and enables customers to solve their AI business requirements by effortlessly deploying models to edge devices. Customers use the SDK to bring their applications into the Deepgrid AI platform and run it on our powerful AI Processing Units (AIPUs).
Whether the application is developed using proprietary or standard industry models, the DGrid SDK offers end-to-end integration and is API-compatible with de-facto industry standards, unleashing the potential of our AIPUs, delivering high-performance AI that can be deployed quickly and easily.
- Your modelTensorFlow Lite · PyTorch · OpenCV · ONNX
- Model OptimizationAutomated quantization and pruning
- Runtime LibrariesHigh-performance inference
- Deepgrid AIPUsAI Processing Units at the edge
- Performance ProfilingAnalyze and optimize model performance
Key Features
Everything you need to develop for Deepgrid platforms
CLI Tools
Command-line interface
Runtime Libraries
High-performance inference
Documentation
Comprehensive guides, API references, and tutorials to get you started quickly.
Read morePerformance Profiling
Built-in profiling tools to analyze and optimize model performance
Pre-trained Models
Model zoo access
Supported Frameworks
Work with your favorite AI frameworks
TensorFlow Lite
PyTorch
OpenCV
ONNX
Ready to Start Developing?
Download the DGrid SDK and start building AI applications for edge devices.