NVIDIA
★ 22,463NemoClaw
透過託管推理，在 NVIDIA OpenShell 內更安全地運行 Hermes、LangChain Deep Agents 和 OpenClaw 等代理程式。
HYSEN LABS 專案目錄
NVIDIA
來自 NVIDIA 組織的可信儲存庫、分類與深度解析。
28 個精選開源專案
NVIDIA
★ 17,309SkillSpector
AI 代理技能的安全掃描器。偵測漏洞、惡意模式和安全風險。
NVIDIA
★ 14,627TensorRT-LLM
TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way.
NVIDIA
★ 13,348TensorRT
NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.
NVIDIA
★ 10,437cutlass
CUDA Templates and Python DSLs for High-Performance Linear Algebra
NVIDIA
★ 9,258garak
the LLM vulnerability scanner
NVIDIA
★ 8,615OpenShell
OpenShell is the safe, private runtime for autonomous AI agents.
NVIDIA
★ 7,117warp
用於 GPU 加速模擬、機器人和機器學習的 Python 框架。
NVIDIA
★ 5,759DALI
A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.
NVIDIA
★ 5,277cuml
NVIDIA cuML: GPU-Accelerated Machine Learning
NVIDIA
★ 5,092nccl
Optimized primitives for collective multi-GPU communication
NVIDIA
★ 4,181GenerativeAIExamples
Generative AI reference workflows optimized for accelerated infrastructure and microservice architecture.
NVIDIA
★ 3,810Model-Optimizer
量化、蒸餾、剪枝、神經架構搜尋、推測解碼等 SOTA 模型最佳化技術的統一函式庫。它為 TensorRT-LLM、TensorRT、vLLM 等下游部署框架壓縮深度學習模型，以優化推理速度。
NVIDIA
★ 3,533TransformerEngine
A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.
NVIDIA
★ 3,301skills
專案速覽：適用於 NVIDIA 產品的代理技能，安裝到 Claude Code、Codex 和其他編碼代理程式中，以端對端運行實體 AI、機器人、模擬、CUDA 和 RAG 工作流程。技能目錄產品|說明 |技能 |愛智商 | NVIDIA AI-Q 藍圖 - 部署本地 AI-Q 服務並作為代理技能運行淺層或深層研究工作流程。
NVIDIA
★ 3,253physicsnemo
Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods
NVIDIA
★ 2,632NeMo-Agent-Toolkit
The NVIDIA NeMo Agent toolkit is an open-source library for efficiently connecting and optimizing teams of AI agents.
NVIDIA
★ 1,209kvpress
LLM KV cache compression made easy
NVIDIA
★ 1,131earth2studio
用於探索、建構和部署人工智慧天氣/氣候工作流程的開源深度學習框架。
NVIDIA
★ 1,042raft
RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.