# NVIDIA cuDF

> Open-source Python GPU DataFrame library that accelerates popular data engines like Apache Spark, pandas, and Polars on NVIDIA AI infrastructure. Built on Apache Arrow, it utilizes GPU parallelism and memory bandwidth to accelerate data processing and analytics workflows, serving as the data-processing foundation for the Sirius GPU-accelerated database project.

- **URL:** https://github.com/rapidsai/cudf
- **Category:** Data Processing
- **Tags:** GPU Accelerated, dataframe, Apache Arrow
- **Updated:** 2026-04-04 23:01
- **Canonical page:** https://vectordb.ever.works/it/items/nvidia-cudf

## Details

## Overview

NVIDIA cuDF is an open-source GPU DataFrame library that provides pandas-like APIs for data processing accelerated by NVIDIA GPUs. Built on Apache Arrow, it utilizes GPU parallelism and memory bandwidth to accelerate data processing and analytics workflows.

## Key Capabilities

- Accelerates popular data engines including Apache Spark, pandas, and Polars
- Built on Apache Arrow columnar memory format
- Utilizes GPU parallelism and memory bandwidth for high-throughput processing
- Native compatibility with DuckDB
- Serves as the data-processing foundation of the Sirius GPU-accelerated database

## Use Cases

- Large-scale data processing and analytics
- Real-time data transformation pipelines
- GPU-accelerated SQL query execution
- Data preprocessing for machine learning and AI workloads

## Licensing

Free and open-source.

---

_This Markdown mirror is generated by Ever Works for AI agents. The canonical HTML page is at https://vectordb.ever.works/it/items/nvidia-cudf._