dgx 开源项目推荐（59 个）

Avarok-Cybersecurity /atlas

纯锈推断引擎

joeynyc /spark-doctor

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.

niklasfrick /spark-dashboard

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

RehmanaYounis /greenray-aio

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

seanGSISG /dgx-spark-sunshine-setup

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.

jdaln /dgx-spark-inference-stack

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

Seanbo5386 /dc-lab-sim

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

Mekopa /whisperx-blackwell

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

timothystewart6 /ubuntu-gb10

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

akaion-ai /annona

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

phuongncn /qwen3.6-27b-speedhack-gx10-dgx-spark

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

riskfuel /k8s-mig-operator

A kubernetes operator for managing nvidia MIG instances.

jef-sure /handwritten-clock

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

0xWhiteMage /qwen3.8-27b-kearuga-sglang-dgx-spark-dflash2

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.

parallelArchitect /cuda-unified-memory-analyzer

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

Ikaran8623 /dgx-spark-ai

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

rootveda /lisa-realtime-voice-assistant

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

eioncustom /DGXDiagOS

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

ondigo-winder /SWITCHLESS-NCCL-RING-4-GB10

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.

ArchitGargPro /Pedestrian-Detector

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.