nopnoping / micro-vllm
A small, educational vLLM implementation for exploring LLM inference, scheduling, KV-cache management, and GPU execution.
nopnoping / micro-vllm
bt420000-collab / ComfyUI-H3-MultiGPU
—
SlugLab / HBFSim
HBFSim (HBF Sim): an open-source High-Bandwidth Flash (HBF) simulator. Runs real LLM inference on a real GPU and injects modeled HBF timing, capacity and thermal effects while the workload executes.
recastsystems / woof
Recast WOOF: GPU weather model (regional, hex and global)
TKY-27 / iris-3b-metal
Iris-3B pixel-space text-to-image on Apple Silicon GPUs: Metal-optimized engine, native macOS app, CLI, Python and local HTTP API
Amduraznak / minimax-h3-fp16-fix
Run MiniMax H3 at native fp16 on pre-bf16 GPUs (V100/Volta) — exact-math fix for the black-frame overflows, ~11× faster than the fp32 fallback.
Rewsr / unawesome-ai-fabric-engineering
Curated resources on RDMA, collectives, and AI cluster fabrics for GPU performance engineers
Jinlong-cs / pi.cpp
TensorRT-first robot foundation model inference for edge GPUs
JMS1717 / Quest3-Pyrowave
Open-source Quest 3 PCVR streamer (ALVR fork) with PyroWave wavelet video decoded in Vulkan on the Quest's GPU. Quality 120 Hz over USB out of the box; tested profiles picked in the headset.
telekinesis-ai / telekinesis-rlbotics
Lightweight, GPU-accelerated reinforcement learning for Isaac Lab, mjlab, and Gymnasium.
ormandj / sglang-glm53-flash-sm120
SGLang for GLM-5.3-Flash on two or four RTX PRO 6000 Blackwell GPUs (SM120), with TP2 W4A16 and TP4 NVIDIA NVFP4 profiles.
arpanpathak / gpu-parallel-book
📖 GPU parallel programming from the ground up : CUDA C++ and Rust 🦀 kernels paired with clickable SVG 2D/3D architecture diagrams (warps, SMs, memory tiers). Includes reduction, scan, matmul, and a full image‑processing capstone
michaelc-yu / cs336-llm-systems
Large language model systems implementation and optimization: Transformers, BPE tokenization, FlashAttention, Triton, distributed training, GPU profiling, and GRPO reasoning RL
PedroMarinhoDev / ComfyUI-HunyuanImage3
Run Tencent's HunyuanImage-3.0 (80B MoE) natively in ComfyUI on a 24 GB GPU: base, Instruct and Instruct-Distil, prompt rewriting, image editing, Spectrum speed-up.
EXwhitewood / wanxiangpu-loreweft
万象谱 Loreweft｜面向长篇小说工程的 AI 原生创作工作台｜AGPL-3.0 开源
jceronch1 / open_suno
Programa para crear canciones tanto con GPU como CPU
ECLIPSE-AI4Science / gradsolve
Differentiable ensemble solvers for differential equations on GPUs, in JAX
sullivanashby3 / runpod-docs
GPU API documentation patterns covering quickstarts, API references, troubleshooting, MCP, examples, and reproducible developer workflows.
mini-yifan / open-source-video-gen-skill
Vibe Video：一句话想法 → 成片。开源 MiniMax H3 + AutoDL 弹性 GPU + Agent 自动开关机，10 秒 1080p ≈ ¥0.3。10 个技能组成的 AI 剧组（编剧/美术/导演/配乐/配音），短剧、宣传片、创意视频都能做。
nibor1896 / Crow
Crow — a local AI agent, not a chat box: 30 tools, MCP, memory, skills, goals, subagents, vision, and image generation and editing with Qwen-Image. Runs on crow-nest or llama.cpp on one GPU, or any model via OpenRouter. Windows and Linux.
PyModel / watermark-remover
Find and remove AI provenance signals from files you own: hidden Unicode, token watermarks, visible image marks, and C2PA/EXIF/XMP metadata. Zero-dependency Python 3.10+ core; models, GPUs, and system binaries sit behind opt-in adapters.
haha-yuki-haha / AUGUstus
🚗 AUGUstus: Autonomous Driving VLA Model with Unified Generation and Understanding. 🏋️ 512 H800 GPUs · ~2 days · 318K+ video hours · 513B+ training tokens. ⚡ 12 Hz on-vehicle deployment. SFT + GRPO + NAVSIM.
wieslawsoltes / Performance-Skill
Cross-platform .NET performance engineering skill for coding agents, covering CPU, memory, GC, benchmarking, concurrency, startup, native profiling, GPU rendering, and production diagnostics on macOS, Windows, and Linux.
buqi-code / buqi-minimax-h3-multigpu
MiniMax-H3 multi-GPU parallel inference for ComfyUI | 多卡并行加速节点:2-8 GPU Ulysses sequence parallel, bit-identical video+audio generation
drowzeys / keys-SM121-Optimized-MiniMax-H3-Nvidia-Sol-Engine-Kijai-SolAttn_Triton-Single-DGX-Spark
MiniMax-H3 33B audio+video DiT accelerated 1.54x on a single NVIDIA DGX Spark (GB10, sm_121): NVIDIA Sol-Engine FirstBlockCache + batched VAE decode ported to single-GPU ComfyUI, Sol-Attn patched for compute 12.1, kijai Triton INT8/TMA kernels. Benchmarks, charts, one-shot installer, agent recipe.
hbofz / NeSLE
GPU-native NES emulator + RL stack: thousands of parallel NES envs on one CUDA GPU, training agents on Super Mario Bros.
cair-vinuni / FoldQuantVLA
FoldQuant: offline folding for W4A4 / W8A8 vision-language-action inference on edge GPUs
ExVRAM / exvram-lab
Compute-for-memory research lab for running larger LLMs on memory-constrained consumer GPUs. Official website exvram.pro
xiaohui5206 / let-ai-read-video
让任何 AI 真正"看"视频：纯本地的视频阅读 skill——faster-whisper GPU 转写 + ffmpeg 场景感知抽帧，双通道带时间戳（t=MM:SS），数据不出机。1 小时视频 3~5 分钟读完。Give any AI agent eyes for video: 100% local video-reading skill — faster-whisper GPU transcription + scene-aware ffmpeg frame extraction, dual-channel with t=MM:SS citations. Nothing leaves your machine.
ArcInstitute / cell-eval2
GPU-accelerated evaluation of perturbation prediction models
kadenball / qwen38-27b-rtx3060-dcfr
RouteWeaver: experimental cache-safe speculative inference and CPU/GPU transfer optimization for Qwen 27B on low-VRAM NVIDIA GPUs.
SoundsGoodAI / fast-gpu-asr
High-throughput batched ASR for NVIDIA GPUs. Run Zipformer and NVIDIA NeMo Parakeet models through one Python API with TensorRT, custom CUDA plugins, GPU beam search, and word timestamps - up to 25,000 RTFx on B300.
dg1kjd / sglang-sxm2
Agentic-coding oriented inference engine for N x V100-SXM2, based on SGLang: GLM-5.3-Flash, Qwen3.8-Flash-Next, DeepSeek-V4.1-Flash, MiniMax-H3.
khoinguyen88kt / chrome-macpro-gpu-patch
Restore full Google Chrome hardware acceleration (OpenGL, WebGL, WebGPU) on legacy Mac GPUs running macOS Sequoia & Sonoma via OCLP
hui-tony-zk / h3zero
MiniMax H3 video generation with zero idle GPU cost.
Fenon-Robotics / fgate
GPU-native quality gate for first-person video
tonywestonuk / YuE-Studio
YuE Studio: YuE2 music generation as a native Mac app, tuned for Apple Silicon (GPU + Neural Engine pipeline)
nisesen / inferenceinquire
Real-time monitoring for llama.cpp inference servers: a read-only host agent plus a web dashboard for hardware, health and inference telemetry (prefill/decode vs context depth, slot occupancy, KV cache, GPU and PCIe).
eust-w / amd-bigym-3dgs-rocm
AMD Radeon ROCm reproduction of the DL3DV commercial-kitchen 3DGS shell for BiGym/MuJoCo, with verified W7900D artifacts.
adnogpu / TXCaptcha
tx防水墙验证码自动化求解器：滑块、图形点选、文字点选三种类型。Python + OpenCV + ONNX 实现，curl-cffi 模拟 Chrome TLS 指纹。仅供学习用途。
scottstts / Friends-Apartment
A procedural Three.js WebGPU recreation of Monica's apartment from Friends.
bennyborn / xbox-boot-re
The boot animation of the original Xbox reconstructed from the retail BIOS alone, and rebuilt from that reconstruction as a standalone renderer with a WebGPU front end.
c8dhjp4tyv-bit / hawkpoint-npu-llm
Run LLMs and Ollama Qwen across AMD XDNA1 NPU, CPU, and GPU on Ryzen AI Linux systems.
arczhi / 5060ti-qwen3.8-27b
Qwen3.8 Local Coding on a 16GB GPU
Tenosra / Proxima
Proxima lets existing GPUs serve 4x more concurrent requests
intel / skills
Skills for running, porting, profiling, and tuning code on Intel CPUs, GPUs, and the libraries and runtimes behind them.
exaloop / codon
A high-performance, zero-overhead, extensible Python compiler with built-in NumPy support
cupy / cupy
NumPy & SciPy for GPU
keylase / nvidia-patch
This patch removes restriction on maximum number of simultaneous NVENC video encoding sessions imposed by Nvidia to consumer-grade GPUs.
google-deepmind / mujoco_playground
An open-source library for GPU-accelerated robot learning and sim-to-real transfer.
google-deepmind / mujoco_warp
GPU-optimized version of the MuJoCo physics simulator, designed for NVIDIA hardware.
NVIDIA-BioNeMo / bionemo-recipes
BioNeMo Recipes: For building and adapting AI models in drug discovery at scale
omnichar / OmniChar
The open .char format for consistent characters. Face, body, wardrobe & voice in one file, consistent across every model.
pytorch / pytorch
Python library providing GPU-accelerated tensor computation like NumPy, plus a tape-based autograd system for building deep neural networks.
meta-pytorch / dist_moe
Dist-MoE implements Speed-of-Light distributed Mixture-of-Experts forward and backward kernels. They are highly optimized and fused comm-compute kernels for training and inference on SM100+ GPUs
crit-labs / crit
Why wait for someone to optimize an LLM for your GPU when an AI agent can build a specialized inference engine in a few hours?
runzhliu / aik8s
AI/LLM on Kubernetes 基础设施知识库：涵盖 GPU、大数据、调度、分布式训练、LLM 推理、RAG、Agent 与生产运维。