Build a voice-interactive AI chatbot on Jetson for physical AI applications using Ollama (LLM), NVIDIA Riva (STT/TTS), and Docker. Fully offline voice assistant with real-time speech recognition and synthesis. Requires JetPack 6.0+ and NGC API key.
Run local text-to-image generation on Jetson Orin NX 16GB using Stable Diffusion. Covers three approaches — KerasCV, Hugging Face diffusers, and NVIDIA Jetson Containers (AUTOMATIC1111 WebUI). Includes a Flask API for on-demand image generation. Requires JetPack 5.1.1+.
Deploy a local RAG chatbot on Jetson using LlamaIndex + ChromaDB + quantized Llama2-7b (MLC). Uses jetson-containers Docker environment. Requires Jetson with ≥16GB RAM and JetPack 5.1+.
Install and configure Lumeo no-code video analytics gateway on NVIDIA Jetson. Covers account creation, gateway deployment, camera linking, and building a people detection pipeline. Requires Jetson with JetPack 5.1+ and internet access.
Deploy real-time YOLO11 object detection and VGGT 3D reconstruction on Jetson AGX Orin with up to 8 GMSL cameras via the reServer Industrial J501 carrier board and GMSL extension board. Requires JetPack 6.2 and CUDA 12.6.
Install and configure NEQTO Engine for Linux on reComputer J30 series. Covers account creation, API key setup, daemon installation, device registration, and Hello World script deployment via NEQTO Console. Requires any Linux-based Jetson with ≥32MB disk and network adapter.
Deploy the No-Code Edge AI Tool (Node-RED based) on reComputer Jetson Nano for drag-and-drop object detection using a live camera. Requires Jetson Nano with JetPack 4.6.1 and a V4L2 USB camera.
Set up NVStreamer for RTSP video streaming on reComputer J4012 with Jetson Platform Services. Covers installation, video upload, RTSP stream creation, and VST integration. Requires JetPack 6.0+ with CUDA 12.2.
Deploy OpenClaw (Clawdbot) locally on reComputer Jetson with Ollama for a self-contained AI control hub. Covers Ollama installation, model pulling, OpenClaw setup, and local model configuration. Requires Jetson with ≥16GB RAM and JetPack 6.2.
Perform NVIDIA Jetson Over-the-Air (OTA) updates on reComputer devices to upgrade from JetPack 5.1.3 to JetPack 6.2 without USB re-flashing. Covers pre-built OTA payloads and custom OTA package generation. Requires Ubuntu x8664 host PC and Orin-based reComputer.
Install the Pinocchio rigid-body dynamics library (stack-of-tasks/pinocchio) into a specified conda environment or a local Python venv. Supports conda (conda-forge), pip, and from-source (cmake) methods. Use when the user asks to install, set up, or build Pinocchio for robotics, MPC, or trajectory optimisation work.
Deploy quantized Llama2-7B with MLC LLM on Jetson Orin NX for fast edge inference. Uses jetson-containers Docker workflow with 4-bit quantization (q4f16ft). Requires Jetson Orin with ≥16GB RAM, JetPack 5.x, and HuggingFace access token.
Deploy a real-time speech-to-subtitle recorder on Jetson using NVIDIA Riva ASR, Flask, and a USB microphone (reSpeaker). Keeps meeting transcription fully on-device for privacy. Requires Jetson with Riva ASR server running.
Deploy Roboflow inference server on NVIDIA Jetson for real-time AI model inference on webcam streams. Supports pip, Docker Hub, and local Docker build methods. Requires JetPack 5.1.1 and a Roboflow API key.
Deploy a Visual Language Model (VLM) AI service on Jetson using Jetson Platform Services (JPS). Requires reComputer J4012 (Orin NX 16GB), JetPack 6.0, and the vlm-1.1.0 package.
Deploy a knife detection model on Triton Inference Server using reComputer J1010 for X-ray security scanning, with Raspberry Pi clients sending images for inference and displaying detection results.
Guidance on safely upgrading software packages on Seeed Jetson devices, explaining why OTA/partial updates are risky and recommending full JetPack upgrades or targeted package installs instead.
Run a multimodal Visual Language Model (VLM) with speech interaction on reComputer Jetson (AGX Orin 64G or Orin NX 16G), combining NVIDIA VLM, SenseVoice speech-to-text, and Coqui-ai TTS for voice-driven visual scene understanding.
Enable SPI1 interface on Jetson Nano by modifying the device tree before flashing. Covers DTB decompilation, SPI pin patching, recompilation, flashing, and loopback testing. Must be applied before flashing the image.
Retrieve boot and system logs from reComputer J30/J40 series devices via the Jetson serial port (J15 header) using a USB-to-TTL adapter and serial debugging tool (PuTTY, XShell, or MobaXterm).
Install NVIDIA-optimized PyTorch with CUDA support on reComputer Jetson devices, covering JetPack 5.1.3 and JetPack 6.2 workflows, cuSPARSELt setup, verification, and managing multiple PyTorch versions with Miniconda.
Deploy an intelligent traffic management system using NVIDIA DeepStream SDK on Jetson devices, with DashCamNet for vehicle detection and VehicleTypeNet/VehicleMakeNet for classification. Tested on reComputer J1010 with JetPack 4.6.1.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: