1. [Home](https://developer.nvidia.com) 2. NVIDIA CUDA # NVIDIA CUDA CUDA is NVIDIA’s platform for accelerated computing and the foundation for GPU computing. Download: [Download CUDA Toolkit](/cuda-downloads "Download CUDA Toolkit") Quick Links: - [Documentation](https://docs.nvidia.com/cuda/) - [Programming Guide](https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html) - [Tutorials and Samples](https://github.com/NVIDIA/accelerated-computing-hub) - [GPU Support](/cuda-gpus) - [Forum](https://forums.developer.nvidia.com/c/accelerated-computing/cuda/206) - [FAQ](/cuda-faq) ## Get Started With CUDA ![NVIDIA CUDA toolkit](https://developer.download.nvidia.com/images/cuda-kv-toolkit-ari.jpg) ### CUDA Toolkit The NVIDIA® CUDA® Toolkit provides the development environment for creating high-performance, GPU-accelerated applications. The toolkit includes GPU-accelerated libraries, debugging and optimization tools, a C++ compiler, and a runtime library. [Get the CUDA Development Environment](/cuda-toolkit) ![NVIDIA CUDA Python logos](https://developer.download.nvidia.com/images/cuda/cuda-python.jpg) ### CUDA Python As one of the most popular programming languages today for AI and high-performance computing (HPC), Python developers can build robust GPU applications directly in Python. [Write GPU-Powered Python](/cuda-python) ![NVIDIA CUDA-X Libraries](https://developer.download.nvidia.com/images/cuda/cuda-tile-1920-1080.jpg) ### CUDA Tile NVIDIA CUDA Tile is the GPU programming model that simplifies the creation of optimized, tile-based kernels and targets portability for special-purpose hardware including Tensor Cores. [Unlock Peak GPU Performance](/cuda/tile) ![NVIDIA Nsight Developer Tools](https://developer.download.nvidia.com/images/cuda/nsight-developer-tools.jpg) ### Nsight Developer Tools NVIDIA Nsight™ tools are a powerful set of libraries, SDKs, and developer tools spanning across desktop and mobile targets. They enable developers to build, debug, profile, and develop software that utilizes the latest accelerated computing hardware. [Build, Debug, and Profile Software](/tools-overview) ![NVIDIA CUDA Tile](https://developer.download.nvidia.com/images/cuda/cuda-x-libraries.jpg) ### CUDA-X Libraries NVIDIA CUDA-X™, built on CUDA, is a collection of libraries that deliver dramatically higher performance across application domains, including AI and HPC. [Explore Prebuilt, Optimized Libraries](/cuda/gpu-accelerated-libraries) ## CUDA Fundamentals [CUDA Programming Guide](https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html) [![NVIDIA CUDA platform for accelerated computing](https://developer.download.nvidia.com/images/cuda/what-is-cuda.svg)](https://developer.download.nvidia.com/images/cuda/what-is-cuda.svg) _Click Image to Enlarge_ ### What Is CUDA? CUDA is NVIDIA's platform for accelerated computing, providing the software layer that enables applications to harness the power of GPUs. Developers can program in languages such as C++, Python, and Fortran or leverage GPU-accelerated libraries and frameworks like PyTorch. This flexibility lets developers integrate GPU computing into any layer of their software stack to achieve optimal functionality and performance. The [CUDA Toolkit](/cuda-toolkit), an integral component of the CUDA platform, provides the compiler, libraries, and developer tools required to develop GPU applications. ### What’s CUDA All About Anyway? Learn about the CUDA ecosystem that helps developers solve real-world challenges. [Watch Video](https://www.nvidia.com/en-us/on-demand/session/gtc25-S72571/) ### Learn CUDA C++ Learn the fundamentals of CUDA C++ with a collection of guided notebooks. [Start Learning](https://github.com/NVIDIA/accelerated-computing-hub/tree/main/gpu-cpp-tutorial) ### Learn CUDA Python Get started with GPU development using Python with a collection of guided notebooks. [Start Learning](https://github.com/NVIDIA/accelerated-computing-hub/tree/main/tutorials/accelerated-python) ### How to Write a CUDA Program Learn about the CUDA ecosystem and how to write CUDA programs. [Watch Video](https://www.nvidia.com/en-us/on-demand/session/gtc25-s72897/) ## Examples of How CUDA Is Used Today ### Artificial Intelligence ### LLM Training Train a reasoning module using NVIDIA NeMo™ Framework and NeMo Curator. - Blog: [Train a Reasoning-Capable LLM in One Weekend With NVIDIA NeMo](https://developer.nvidia.com/blog/train-a-reasoning-capable-llm-in-one-weekend-with-nvidia-nemo/) - Code: [NeMo Framework](https://github.com/NVIDIA/NeMo) - Notebook: [Train Your Own Reasoning Model in 48 Hours](https://github.com/NVIDIA-NeMo/NeMo/blob/main/tutorials/llm/reasoning/Reasoning-SFT.ipynb) ### Artificial Intelligence ### LLM Inference Deploy AI models using NVIDIA Dynamo, an open-source, low-latency, modular inference framework. - Blog: [NVIDIA Dynamo, A Low-Latency Distributed Inference Framework for Scaling Reasoning AI Models](https://developer.nvidia.com/blog/introducing-nvidia-dynamo-a-low-latency-distributed-inference-framework-for-scaling-reasoning-ai-models/) - Guide: [NVIDIA Dynamo](https://github.com/ai-dynamo/dynamo) - Training: [Deploy LLM Inference With NVIDIA Dynamo and vLLM](https://github.com/ai-dynamo/dynamo/tree/main/docs/guides) ### Computer-Aided Engineering ### AI-Powered CAE Simulations Accelerate your CAE simulations with CUDA-X-accelerated CAE tools, AI emulation, GPU acceleration, and real-time digital twins to design and build new technologies. - Blog:[How to Run AI-Powered CAE Simulations](/blog/how-to-run-ai-powered-cae-simulations/) - Guide: [NVIDIA PhysicsNeMo™](https://docs.nvidia.com/physicsnemo/index.html) - Training: [Accelerating Computer-Aided Engineering (CAE) With NVIDIA AI Physics Technology](https://learn.nvidia.com/courses/course-detail?course_id=course-v1:DLI+S-OV-48+V1) ### Data Science ### DataFrame and SQL Acceleration With cuDF cuDF is a GPU-accelerated library that optimizes fundamental DataFrame and SQL operations. It includes drop-in accelerators for popular DataFrame tools like pandas, Polars, and Apache Spark with no code changes required. - Notebook: [Build an Interactive Data Analytics Dashboard](https://colab.research.google.com/gist/will-hill/aa24c3ffe1428c005af3793fcacf9bd2/cudf_pandas_opencellid_demo.ipynb) - Video: [Accelerated Exploratory Data Analysis With pandas on NVIDIA GPUs (16:07](https://www.youtube.com/watch?v=PJpCJsqcfOk)) - User Guide: [cuDF](https://docs.rapids.ai/api/cudf/stable/) - User Guide: [Apache Spark Accelerated With cuDF](https://nvidia.github.io/spark-rapids/) ### Quantum Computing ### Accelerated Quantum Computing With NVIDIA CUDA-Q NVIDIA CUDA-Q™ is the quantum processing unit (QPU)-agnostic platform for accelerated quantum supercomputing. - Blog: [NVIDIA CUDA-Q Powers Quantum Applications Research](/blog/nvidia-cuda-q-powers-quantum-applications-research/) - Docs: [CUDA-Q](https://nvidia.github.io/cuda-quantum/latest/index.html) - Training: [CUDA-Q Academic](https://github.com/NVIDIA/cuda-q-academic) ### Robotics ### World Foundation Models With NVIDIA Cosmos Accelerate physical AI development for autonomous vehicles (AVs), robots, and video analytics AI agents. - Blog: [AI Foundation Models With NVIDIA Cosmos™ Predict-2](/blog/develop-custom-physical-ai-foundation-models-with-nvidia-cosmos-predict-2/) - Github: [Cosmos Predict-2](https://github.com/nvidia-cosmos/cosmos-predict2) - Github: [Cosmos](https://github.com/nvidia-cosmos) [View Accelerated Computing Learning Path](https://nvdam.widen.net/s/brxsxxtskb/dli-learning-journey-2009000-r5-web) ## CUDA Resources Blogs Sessions Training ### An Even Easier Introduction to CUDA An interactive accompaniment to Mark Harris's popular blog post “An Even Easier Introduction to CUDA.” [Learn More About CUDA](https://learn.nvidia.com/courses/course-detail?course_id=course-v1:DLI+T-AC-01+V1) ### Getting Started With Accelerated Computing in Modern CUDA C++ Learn how to write, compile, and run GPU-accelerated code. [Learn More About Accelerating C++ Code with CUDA](https://learn.nvidia.com/courses/course-detail?course_id=course-v1:DLI+S-AC-04+V2) ### Accelerated Computing Hub Learn more about how to use CUDA in the Accelerated Computing Hub, which includes C++ and Python step-by-step tutorials and user guides. [Learn More About Accelerated Computing Hub](https://github.com/NVIDIA/accelerated-computing-hub) [View Accelerated Computing Learning Path](https://nvdam.widen.net/s/brxsxxtskb/dli-learning-journey-2009000-r5-web) ## Get started with CUDA today. [Download CUDA Toolkit](/cuda-downloads "Download CUDA Toolkit")