dgx

Pure Rust Inference Engine

Local diagnostic CLI for NVIDIA DGX Spark (GB10). Detects power caps, UMA pressure, thermal risk, CUDA 13/SM_121 wheel mismatches, Docker issues, and validates vLLM/Ollama/llama.cpp/SGLang recipes.

Real-time hardware and LLM inference monitoring — GPU, CPU, memory, and vLLM metrics streamed to a dashboard.

Hands-on lab simulator and mock exam for the NVIDIA NCP-AIO certification. Simulated DGX H100 cluster, Slurm, Kubernetes, BCM. No GPU needed.

Headless 4K remote desktop for the NVIDIA DGX Spark (GB10): one-command installer for Sunshine + Moonlight low-latency game streaming with NVENC hardware encoding, a software virtual display (no HDMI dummy plug), GDM autologin, and optional Tailscale.

Serve the home! Inference stack for your Nvidia DGX / RTX Spark (128GB ram) aka the Grace Blackwell AI supercomputer on your desk. Mostly vLLM based for now and single-spark. For the not-so-rich people. Updated once a month. Hobby project

Browser-based datacenter lab simulator for NCP-AII certification exam prep — 20 command simulators, 32 guided scenarios, and a full learning progression system.

GPU-accelerated WhisperX on NVIDIA Blackwell (SM_121) - DGX Spark compatible

Ubuntu 24.04 + NVIDIA stack setup guide for GB10 / DGX Spark systems

Annona — the sovereign execution kernel for AI agents. Decides where each step runs, enforces it, and records it.

Qwen3.6 27B × DFlash — 30-35 tok/s on NVIDIA DGX Spark (GB10) - LLama.Cpp

A kubernetes operator for managing nvidia MIG instances.

ESP32 weather clock in a handwritten font of Bézier curves: morphing digits, pen-written text, weather icons flowing in dots

Qwen3.8-27B on one DGX Spark GB10: DFlash 2 for 65 tok/s C1 and 120 C4; EAGLE 3/1/4 for ~191 C8, ~330 C16 and ~535 C32; native 262K, pinned SGLang.

gpu thrashingNVIDIA GPU Unified Memory diagnostic tool — architecture-aware, measurement-based, PCIe/coherent transport detection

Run GPT-OSS 120B on NVIDIA DGX Spark with vLLM, build an API server, and create a local AI coding assistant

Lisa — Real-time Voice Assistant. Offline stack: Nemotron ASR, Gemma 4, XTTS, RAG, vision.

This is a home brewed menu system for Nvidia's DGX Field Diagnostics Suite.

NCCL over a switchless 4-node DGX Spark / GB10 ring using both PCIe halves of every QSFP cable: ~193 Gb/s per cable instead of ~112, up to +34% vLLM prefill. One patch on top of switchless-nccl.

Pedestrian Detector— using Faster RCNN. This is an application of computer vision, object recognition. We detect people walking on the road and cyclists. Research is also done comparing its performance with different models like YOLO v3.