Seeed-Jetson-DevelopTool

Install NVIDIA-optimized PyTorch with CUDA support on reComputer Jetson devices, covering JetPack 5.x and JetPack 6.x workflows, cuSPARSELt setup, torch/torchvision pairing, probing multiple Python or conda targets, and managing multiple PyTorch versions with Miniconda.

Deploy a local AI NVR (Network Video Recorder) on Jetson Orin using NVIDIA VST and Jetson Platform Services with DeepStream pedestrian detection. Requires Jetson Orin with JetPack 6.0 and an IP camera.

Perform over-the-air JetPack/L4T updates on Jetson devices using Allxon DMS Portal. Covers OTA payload generation, upload, and fleet distribution. Requires Ubuntu host PC and Allxon DMS agent.

Install and configure Allxon DMS agent on NVIDIA Jetson devices for remote monitoring and management via the Allxon cloud portal. Supports JetPack 4.6+.

Set up the alwaysAI computer vision platform to deploy ML-based object detection on Jetson devices. Supports live camera and video file inference with TensorRT acceleration. Requires JetPack 4.6 and a host PC.

Backup and restore a reComputer Jetson system image using the L4T backup/restore script. Requires Ubuntu host, USB-C cable, and the matching JetPack BSP package.

Build Seeed's custom BSP source code for Jetson devices — obtain NVIDIA L4T, overlay Seeed patches, compile kernel, and flash. Requires Ubuntu 20.04/22.04 host PC.

Use for Seeed Jetson topic "Develop reComputer Jetson using Clawdbot". Includes local references and execution-safe workflow with prerequisites, commands, and validation checks.

Deploy Cochl.Sense sound recognition on Jetson (37 sounds, 94% F1 score). Requires a trial license from Cochl and a USB microphone. Supports JetPack 4.6+.

Install and run CVEDIA-RT AI inference engine on NVIDIA Jetson for pre-built computer vision applications (crowd estimation, drone detection, vehicle counting, etc.). Requires JetPack 5.1 and a CVEDIA account.

Deploy DashCamNet and PeopleNet pre-trained models on Jetson Xavier NX with multi-camera pipelines using DeepStream, TAO Toolkit, and jetson-multicamera-pipelines. Requires JetPack 4.5/4.6.

Set up the Deci AI platform on NVIDIA Jetson devices (Nano, Xavier NX, AGX Xavier, AGX Orin) for model optimization, benchmarking, and deployment. Covers Deci account creation, INFERY SDK installation, model zoo usage, custom model optimization via AutoNAC, and on-device performance benchmarking with ONNX/TensorRT. Requires JetPack 4.6.

Quickly deploy DeepSeek-R1 on Jetson using Ollama. Two-command setup. Requires Jetson with >8GB RAM and JetPack 5.1.1+.

Deploy DeepSeek on Jetson Orin using MLC (Machine Learning Compilation) for optimized edge inference. Uses Docker/jetson-containers. Requires Jetson with >8GB RAM and JetPack 5.1.1+.

Deploy Depth Anything V3 monocular depth estimation on Jetson AGX Orin with ROS2 Humble integration. Builds a ROS2 workspace with TensorRT-optimized ONNX models for real-time depth map generation from USB camera or video input. Requires JetPack 6.2 and ROS2 Humble.

Deploy the Dia neural text-to-speech model on reComputer Jetson for expressive, multi-speaker audio generation. Installs custom aarch64 PyTorch/torchaudio/triton wheels, clones the Dia repo, patches pyproject.toml, and launches a Gradio WebUI. Requires Jetson with 8GB+ RAM and JetPack 6.1+.

Use for Seeed Jetson topic "Efficient Multi-Task Vision Inference Engine Deployment on Jetson". Includes local references and execution-safe workflow with prerequisites, commands, and validation checks.

Deploy Frigate open-source NVR with real-time TensorRT object detection on NVIDIA Jetson (Nano, Xavier NX, Orin). Uses Docker Compose with the Jetson-optimized Frigate image, YOLOv7 model, and RTSP IP camera streams. Requires JetPack 5.x and Docker.

Deploy Live VLM WebUI on reComputer Jetson for real-time Vision Language Model interaction via webcam. Installs Ollama with llama3.2-vision model and the live-vlm-webui Python package for browser-based VLM streaming and benchmarking. Requires JetPack 6.2 and USB camera.

Deploy Isaac ROS NVBlox with Orbbec RGB-D camera on Jetson AGX Orin for real-time GPU-accelerated 3D mapping (TSDF/ESDF). Sets up Docker CE, Isaac ROS 3.2, Orbbec SDK ROS2, and builds the NVBlox workspace for obstacle detection and 3D mesh generation. Requires JetPack 6.2 and ROS2 Humble.

Use for Seeed Jetson topic "Local AI Assistant". Includes local references and execution-safe workflow with prerequisites, commands, and validation checks.

Build a local voice chatbot on Jetson AGX Orin using NVIDIA Riva (ASR + TTS) and Meta Llama2 via text-generation-inference. Sets up NGC, Riva server, Llama2-7b-chat, and a Python voice chatbot demo with speaker/microphone. Requires JetPack 5.1.1+ and 16GB+ RAM.

Encrypt the root filesystem on Jetson devices during flashing using OPTEE-based encryption keys. Covers BSP source build, OPTEE key generation, encrypted image creation, and flashing with l4t_initrd_flash. Requires an Ubuntu 20.04/22.04 host PC and the target Jetson device (e.g. reComputer J401 with L4T 36.4.3).

Create a custom flashable BSP (Board Support Package) by cloning a complete Jetson development environment. Covers working directory setup, environment cloning via l4t_backup_restore, BSP packaging, and flashing to target devices. Requires Ubuntu 22.04 host PC and USB-C cable.

Establish EtherCAT communication between Jetson and EtherCAT slave devices using the EtherLab EtherCAT Master driver. Covers driver installation, configuration, slave scanning, and motor control example (MyActuator X4). Requires JetPack 6.2 (L4T 36.4.3).

Fine-tune LLMs on Jetson using Llama-Factory via jetson-examples one-line deployment. Covers installation, WebUI-based training with alpaca_zh dataset on Phi-1.5, and testing the fine-tuned model. Requires Jetson with ≥16GB RAM.

Fix browsers (Chromium, Firefox, etc.) that fail to launch on Jetson devices due to an outdated or broken snapd. Downloads a pinned snapd revision offline, installs it, and holds automatic updates to prevent regression. Use when the user reports that a browser installed via snap cannot open, crashes silently, or shows snap-related errors on Jetson (Ubuntu/L4T).

Deploy Gapi embeddable API gateway on Jetson Orin for streaming AI micro service integrations. Covers Docker-based installation, workflow engine setup, and community micro services. Requires any Jetson Orin with Docker and 1.3GB storage.

Overview and index of Generative AI capabilities on reComputer Jetson. Covers text generation (LLM chatbots), image generation, audio generation (Whisper STT), multimodal (VLM), RAG, LLM fine-tuning, and quantized inference. This is a reference/navigation skill — it links to specific deployment skills.

Deploy GPT-OSS-20B on Jetson Orin using llama.cpp with CUDA. Builds llama.cpp from source, converts and quantizes the model to Q4_K, and runs inference. Optional OpenWebUI frontend. Requires Jetson Orin with ≥16GB RAM.

Fine-tune NVIDIA Isaac GR00T N1.5 for the LeRobot SO-101 arm and deploy on Jetson AGX Thor. Covers Thor flashing, dev environment setup, SO-ARM data collection, cloud training via NVIDIA Brev, and GR00T N1.5 inference on Thor. Requires Jetson AGX Thor and SO-101 arm hardware.

Fine-tune NVIDIA Isaac GR00T N1.6 for the LeRobot SO-101 arm and deploy on AGX Orin 64G. Covers environment setup on AGX Orin and fine-tuning server, SO-ARM data collection, dataset format conversion, L20 GPU training, and inference deployment. Requires JetPack 6.2 and ≥48GB VRAM for fine-tuning.

Train and deploy a hard hat detection ML model using Edge Impulse on Jetson Nano/NX/AGX. Covers Edge Impulse project setup, data collection (public datasets, PC camera, or Jetson camera), model training, and deployment via Edge Impulse CLI or Linux Python SDK.

Install IMX477 camera driver on reComputer A603 carrier board. Downloads the JetPack-specific BSP with camera driver included and reflashes the device. Supports JP5.1.2, JP6.0, and JP6.2.

Build a fruit sorting system using J501 Mini (Jetson AGX Orin) with StarAI Viola robotic arm and LeRobot framework. Covers environment setup, arm calibration, teleoperation, data collection, ACT policy training, and autonomous deployment. Requires JetPack 6.2.1.

Flash JetPack on Jetson from Windows using WSL2 + usbipd-win. Covers custom WSL kernel setup, USB passthrough, and Seeed BSP massflash. Not officially supported — use native Ubuntu if issues persist.

Knowledge base for the relationship between JetPack and Jetson — JetPack components (L4T + SDK), Ubuntu version mapping, supported Seeed products, Super Mode, and JetPack archive links.

Perform an image-based OTA update on Seeed Jetson devices to upgrade from JetPack 5.1.3 to JetPack 6.0. Requires Ubuntu 20.04 host PC and USB-C cable. Demonstrated on reComputer J3010.

Fix JetPack 5 boot failures from NVMe SSDs on Seeed Jetson devices. Two options — full SSD wipe via external enclosure, or re-flash with --erase-all flag. JetPack 6 is not affected.

Knowledge base for Jetson AI developer tools and Seeed ecosystem partners — CVEDIA-RT, Lumeo, alwaysAI, YOLOv5/v8, Roboflow, Allxon, Deci, Edge Impulse, Cochl.Sense. Use to recommend tools or find getting-started guides.