# Raspberry Pi AI Hat Model Trainer **URL:** \ **Category:** Projects\ **Created:** [10 September 2026 10:56 UTC](https://forum.core-electronics.com.au/t/raspberry-pi-ai-hat-model-trainer/24424 "2026-09-10T10:56:29Z")\ **Posts on this page:** 1\ **Page:** 1
**Author:** ![Shaun21504](https://yyz2.discourse-cdn.com/flex028/user_avatar/forum.core-electronics.com.au/shaun21504/32/8473_2.png) [@Shaun21504](https://forum.core-electronics.com.au/u/Shaun21504)\ **Post date:** [10 September 2026 10:56 UTC](https://forum.core-electronics.com.au/t/raspberry-pi-ai-hat-model-trainer/24424/1 "2026-09-10T10:56:29Z")
**Pi Trainer: an open-source workbench for Raspberry Pi AI HAT models** ![pi-trainer-forum-header](https://canada1.discourse-cdn.com/flex028/uploads/core_electronics/original/2X/9/9b55b945965fc1bdce7e9d32e56ef1b215a2e807.jpeg) Hi everyone, I’ve been building **Pi Trainer** , an MIT-licensed workbench to help manage the steps between collecting training data and running a model on a Raspberry Pi AI HAT. The idea is to train on a **Mac, Windows PC or Linux machine** , then compile for the appropriate Hailo accelerator, package the model and scripts, and test on the Pi. Training stays off the Raspberry Pi. **GitHub:** [GitHub - ShaunPrice/raspberry-pi-ai-trainer: Cross-platform training, optimisation and deployment workbench for Raspberry Pi Hailo-8L, Hailo-8 and Hailo-10H accelerators. · GitHub](https://github.com/ShaunPrice/raspberry-pi-ai-trainer) ![pi-trainer-web-workbench](https://canada1.discourse-cdn.com/flex028/uploads/core_electronics/original/2X/8/8ca628d7777f03943f250d387db9f0859777c100.jpeg) **Application interfaces** **Web workbench:** the project and dataset view above. **Native desktop:** the same workbench is also available as a desktop application. ![pi-trainer-native-desktop](https://canada1.discourse-cdn.com/flex028/uploads/core_electronics/original/2X/8/8e9c96cdfb572ba2a019a21fd1d7fee4883d73b5.jpeg) **Training workflow:** select a provider, review the settings and submit a background job. ![pi-trainer-training](https://canada1.discourse-cdn.com/flex028/uploads/core_electronics/original/2X/6/6e7a7b7eb85d8440e9d1a7572cd66891a8a30ec7.jpeg) The CLI supports repeatable commands, while MCP exposes the workbench to a connected assistant over stdio. Actual CLI help output: ```auto $ python3.11 -m pi_trainer --data-dir /private/tmp/pi-trainer-demo/store --help usage: main.py [-h] [--data-dir DATA_DIR] {capabilities,projects,jobs,mcp,desktop,web,create,import,plan,bundle,datasets,llm-recipe,score,estimate,workflow-jobs,doctor,job,logs,cancel,run,invoke} ... Pi Trainer command line. positional arguments: {capabilities,projects,jobs,mcp,desktop,web,create,import,plan,bundle,datasets,llm-recipe,score,estimate,workflow-jobs,doctor,job,logs,cancel,run,invoke} options: -h, --help show this help message and exit --data-dir DATA_DIR ``` **What it does** - Provides a native desktop application, web interface, CLI and MCP interface that share projects and jobs. - Imports datasets, records snapshots and hashes, and supports versioned labels and source groups. - Runs image-classification training with PyTorch, exporting ONNX and calibration data. - Compares a bounded set of training configurations to help balance model quality and resource use. - Connects to a compatible Hailo SDK environment for target-specific compilation. - Packages models and scripts with checksums and target/runtime metadata. - Supports SSH deployment and bounded Pi tests, with release selection and rollback controls. - Adds a model bundle and scripts to a **copy of an existing Raspberry Pi OS image**. It does not build a new distribution or resize partitions. The compiler needs a compatible **Linux x86-64 Hailo SDK**. Training frameworks, vendor SDKs and Pi runtimes are separate dependencies. ![pi-trainer-image-workflow](https://canada1.discourse-cdn.com/flex028/uploads/core_electronics/original/2X/4/4a0e24bff352ad228b25d7d0a757594b8c0abf30.jpeg) _\*The screenshot shows image-workflow preparation, with completed training and packaging jobs below it.\*_ **Hardware and testing** Tested on **Raspberry Pi 5 (8GB)** with: - **Raspberry Pi AI HAT+ 13 TOPS — Hailo-8L** - **Raspberry Pi AI HAT+ 26 TOPS — Hailo-8** Synthetic vision compilation and SDK numerical emulation have also passed for **Hailo-10H**. I don’t have the 10H board yet, so its physical testing remains pending. These fixture tests establish pipeline operation, not real-world model accuracy. The full deployment-helper activation and rollback sequence also needs separate hardware qualification. **What about LLMs?** There is a Hailo-10H workflow for host-side LoRA fine-tuning and preparation. Deployment requires a supported Hailo base model, matching pre-optimised archive and a qualified compiler recipe. It is not an arbitrary Hugging Face or GGUF-to-HEF converter. Hailo-8L and Hailo-8 are vision targets in this workbench. ![pi-trainer-llm-preparation](https://canada1.discourse-cdn.com/flex028/uploads/core_electronics/original/2X/f/fe55e05ea8ccaaff0bd8721029f1103b35faab18.jpeg) _\*LLM configuration screen; this screenshot does not show a completed Hailo-10H deployment.\*_ **Guide and walkthrough** I’ve put together a **19-page illustrated guide** \*and a **7-minute narrated walkthrough** showing actual project creation, data import, training and packaging. The compilation, image, device and LLM sections explain preparation and requirements. Watch the Pi Trainer walkthrough on YouTube [![pi-trainer-youtube-thumbnail](https://canada1.discourse-cdn.com/flex028/uploads/core_electronics/original/2X/7/7e8d705b3d345bc8f231a346c818535f2191d44c.jpeg)](https://youtu.be/O7UQfUIOpUI) Watch on YouTube: [https://youtu.be/O7UQfUIOpUI](https://youtu.be/O7UQfUIOpUI) - Read the illustrated PDF guide: [https://github.com/ShaunPrice/raspberry-pi-ai-trainer/releases/download/tutorial-v1/Pi-Trainer-User-Guide.pdf](https://github.com/ShaunPrice/raspberry-pi-ai-trainer/releases/download/tutorial-v1/Pi-Trainer-User-Guide.pdf) - Source code and setup instructions: [GitHub - ShaunPrice/raspberry-pi-ai-trainer: Cross-platform training, optimisation and deployment workbench for Raspberry Pi Hailo-8L, Hailo-8 and Hailo-10H accelerators. · GitHub](https://github.com/ShaunPrice/raspberry-pi-ai-trainer) This is an early workbench, and I’d welcome feedback from anyone using the Raspberry Pi AI HAT range. Which part of your model workflow is the most awkward: dataset preparation, compilation, packaging or testing on the Pi? If you try it, please share your board, runtime version and what worked or failed. _\*Application images are actual captures. Header and thumbnail are AI-assisted promotional artwork based on an [official Raspberry Pi product photograph_]([https://www.raspberrypi.com/products/ai-hat/](https://www.raspberrypi.com/products/ai-hat/)_), credited to Raspberry Pi Ltd; they are illustrative hardware images, not photographs of the test devices. No affiliation or endorsement is implied.\*_