CUDA-X for Data Science
CUDA-X™ is a collection of highly optimized, domain-specific libraries built on CUDA™ that includes a suite of open source libraries for accelerated data science. With 100+ integrations with open source libraries and tools in the data science ecosystem and zero-code-change APIs that accelerate popular PyData tools like pandas and scikit-learn, data scientists can easily accelerate their workflows with their existing tools.
CUDA-X Libraries for Data Science
CUDA-X libraries accelerate data and graph analytics, machine learning, and data visualizations. Data scientists can optimize for performance on single GPUs or scale up to distributed systems.
cuDF: 20x Faster Polars
cuDF is a toolkit that contains GPU-accelerated libraries to optimize fundamental DataFrame operations. It includes drop-in accelerators for popular DataFrame libraries and SQL engines, like Polars, pandas, and Apache Spark with no code changes required.
Learn More About cuDF View DocsTAGS: pandas, polars, apache spark, dataframe, Python, C++
cuML: 50x Faster Scikit-learn
cuML is a GPU-accelerated machine learning library that optimizes machine learning algorithms for execution on GPUs. It includes accelerators that run machine learning algorithms in scikit-learn, UMAP, and HDBSCAN with no code changes required.
View DocsTAGS: scikit-learn, machine learning, Python, C++
cuGraph: 48x Faster NetworkX
cuGraph is a GPU-accelerated graph analytics library that optimizes graph algorithms for execution on GPUs to process millions of nodes without specialized software. It includes a zero-code-change accelerator for NetworkX.
View DocsTAGS: NetworkX, graph, Python, C++
cuxfilter
Create interactive data visuals with multidimensional filtering of over 100-million-row tabular datasets.
Tags: dashboards, visualization, Python
Dask
Scale out GPU-accelerated data science pipelines for machine learning, XGBoost, and graph analytics to multiple nodes on Dask.
Tags: distributed computing, Python
Apache Spark
Accelerate Apache Spark data processing workflows on NVIDIA GPUs with the RAPIDS™ Accelerator for Apache Spark.
TAGS: distributed computing, data processing, Python
Other CUDA-X Libraries for Data Science
See a complete list of libraries and tools.
Install and Deploy in Your Environment
Quick Install With conda
1. If not installed, download and run the install script. This will install the latest miniforge:
wget "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-$(uname)-$(uname -m).sh" bash Miniforge3-$(uname)-$(uname -m).sh
2. Then install with:
conda create -n rapids-26.06 -c rapidsai -c conda-forge rapids=26.06 python=3.14 'cuda-version>=13.0,<=13.2'
Quick Install With pip
pip install \ --extra-index-url=https://pypi.nvidia.com \ "cudf-cu13==26.6.*" \ "dask-cudf-cu13==26.6.*" \ "cuml-cu13==26.6.*" \ "cugraph-cu13==26.6.*"
Deploy Locally
Use this guide to install and build with conda, pip, Docker, or WSL2 on your local machine.
Deploy on Platforms
Deploy on your platform of choice, including Kubernetes, Databricks, and Google Colab.
Data Science Learning Library
The Accelerated Data Science Ecosystem
Data practitioners in open source libraries, commercial software, and industries are driving innovation with CUDA-X.
We're committed to simplifying, unifying, and accelerating data science for the open-source community.
Use CUDA-X libraries in the most popular data science and machine learning platforms.
Industry leaders are driving innovation with CUDA-X.
bunq improved fraud detection accuracy by accelerating model training 100x and data processing 5x using NVIDIA cuDF and cuML libraries.
Capital One accelerated its financial and credit analysis pipelines with NVIDIA cuDF and cuML, improving model training by 100x.
Checkout.com accelerated their data analysis workflows from minutes to seconds with NVIDIA cuDF.
LinkedIn developed DARWIN to enable faster data analysis on NVIDIA cuDF.
TGen cut analysis time on 4-million-cell datasets from 10 hours to three minutes with RAPIDS-singlecell, built on NVIDIA cuML.
Join the Community
Ethical AI
NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their supporting team to ensure their application meets requirements for the relevant industry and use case and addresses unforeseen product misuse.
Please report security vulnerabilities or NVIDIA AI Concerns here.