jetson-thor

FlashRT is a high-performance realtime inference engine for small-batch, latency-sensitive AI workloads. The flagship integration is production VLA control for Pi0, Pi0.5, GROOT N1.6, and Pi0-FAST. Also support llm e.g, qwen3.6-27B

AI agents running research on single-GPU nanochat training automatically for Jetson Thor

Jetson AGX Thor 端侧 AI 与具身智能实战课程：CUDA/TensorRT、ROS 2、视觉感知、坐标变换与安全任务规划。

Lightweight Rust TUI system monitor for NVIDIA Jetson and Linux. Track CPU, RAM, GPU temps, network, disk I/O, and processes.

THOR — Native macOS control center for NVIDIA Jetson devices. Manage power, Docker, ROS2, ANIMA AI modules, cameras, GPIO, and more. No terminal needed.

Measured sm_110 / sm_110a facts for NVIDIA Jetson AGX Thor: tensor-core capability matrix (2:4 sparsity works on plain sm_110; tcgen05 + NVFP4 block-scale is sm_110a-only), CUDA 11/12->13 migration breaks, CPU ISA, and the bandwidth roofline. With probes.

NVIDIA Jetson Thor T5000 感測器影像調適繁中教學站 — 手把手感測器初始化（MIPI/HSB）→ streaming → 調校的 16 單元深度實作教學（參考 T4000，含 Windows/macOS/Linux 開發環境）。Jetson Thor T5000 sensor & ISP tuning teaching site (Traditional Chinese).

말로 지시하면 SO-101 양팔 로봇이 쓰레기를 모으는 Physical AI 데모 — Vision(YOLO) · Language(LLM 의도 게이트) · Action(LeRobot ACT) 통합

Docker packaging to run antirez/ds4 (DwarfStar / DeepSeek-V4-Flash) on NVIDIA aarch64 unified-memory boxes - tested on Jetson AGX Thor (sm_110 , CUDA 13), persistent disk KV cache.

Minimal cuMotion + nvblox + Intel RealSense stack for Jetson AGX Thor (Isaac ROS 4.5 / ROS Jazzy). Perception + planning only.

AI drawing robot: a 4-DOF Waveshare RoArm-M2-S draws with a felt pen on a vertical board, driven from a Jetson AGX Thor — backlash-compensated IK, closed-loop encoder verification, and wrist-cam visual servoing.

Real-world voice AI model benchmarks on NVIDIA Jetson Thor (JetPack 7.0, Blackwell GPU, 122GB unified RAM). Comparing STT, TTS and omni models for low-latency voice assistants.

Container packaging for HART (mit-han-lab/hart) text-to-image generation on NVIDIA aarch64 CUDA / unified-memory boxes - tested on Jetson AGX Thor (sm_110), includes a Blackwell compute-capability fix for the kernel build.