Welcome to DEEPCRAFT™ Documentation
Everything you need to build intelligent and low-power Edge AI.
DEEPCRAFT™ Studio
Build your own model
Build, train, and deploy Edge AI models
- End-to-end ML: from data collection to on-device deployment
- Graph-based UX: visual, node-based workflow for unique insight into ML development process
- Build custom models: develop Edge AI with any time series or image data for any MCU
DEEPCRAFT™ Model Converter
Bring your own model (BYOM)
Convert, optimize and validate AI models to run on Infineon MCUs
- Convert: supports many different AI model file formats
- Optimize: deploy with ease onto Infineon hardware
- Validate: verify performance before deployment
DEEPCRAFT™ Ready Models
Buy a production-ready model
Take Edge AI to market fast with fully trained and tested models
- No ML skills required: complete, validated models built by expert engineers
- Edge-optimized: low power, small memory footprint, fast inference
- Test, buy, deploy: test for free, purchase for royalty per device, deploy on your device
DEEPCRAFT™ Studio Accelerators
Kickstart your custom model development
Shorten your development time with pre-trained, open-source deep learning projects you can fine-tune for your use case
- Download and customize: access in DEEPCRAFT™ Studio and adapt to your needs
- Include everything you need: datasets, preprocessing, model architectures, and clear instructions for developing into production-ready Edge AI models
- Learn by doing: hands-on, end-to-end workflow from data to deployment