{"id":8494,"library":"pylibcudf-cu12","title":"PyLibcuDF (CUDA 12)","description":"pylibcudf-cu12 is the Python binding layer for libcudf, a GPU-accelerated DataFrame library that is part of the NVIDIA RAPIDS ecosystem. It provides high-performance data manipulation primitives, primarily used through the higher-level `cudf` library. As of version 26.4.0, it follows a monthly release cadence, aligned with other RAPIDS components.","status":"active","version":"26.4.0","language":"python","source_language":"en","source_url":"https://github.com/rapidsai/cudf","tags":["GPU","dataframes","RAPIDS","CUDA","dataframe","high-performance","data-science"],"install":[{"cmd":"pip install pylibcudf-cu12","lang":"bash","label":"Recommended installation for CUDA 12"}],"dependencies":[{"reason":"Required Python version.","package":"python","optional":false},{"reason":"This specific package variant requires CUDA Toolkit 12.x to be installed and available on the system.","package":"cuda","optional":false},{"reason":"Minimum PyArrow version required by cudf v26.04.00 and later.","package":"pyarrow>=19","optional":false}],"imports":[{"note":"While 'pylibcudf' is the binding library, users typically interact with the higher-level 'cudf' API.","symbol":"cudf","correct":"import cudf"}],"quickstart":{"code":"import cudf\nimport numpy as np\n\n# Create a cuDF DataFrame from a dictionary\ndata = {'col1': np.random.rand(10), 'col2': np.arange(10)}\ngdf = cudf.DataFrame(data)\nprint(\"Original DataFrame:\")\nprint(gdf)\n\n# Perform a simple operation\ngdf['col3'] = gdf['col1'] * 2\nprint(\"\\nDataFrame after operation:\")\nprint(gdf)","lang":"python","description":"This quickstart demonstrates how to create a basic `cudf.DataFrame` and perform a simple element-wise operation, showcasing the primary entry point for using the library."},"warnings":[{"fix":"Upgrade PyArrow: `pip install 'pyarrow>=19'`","message":"Starting with v26.04.00, `pylibcudf` (and by extension `cudf`) requires PyArrow version 19 or higher. Earlier versions will lead to import or runtime errors.","severity":"breaking","affected_versions":">=26.04.00"},{"fix":"Review `cudf` migration guides for alternative approaches to custom row/chunk-wise operations. Often, these can be replaced by UDFs (User Defined Functions) or other optimized `cudf` methods.","message":"The `DataFrame.apply_chunks`, `Groupby.apply_grouped`, and `DataFrame.apply_rows` methods have been removed. They were deprecated in earlier versions (v25.10.00) and removed in v25.12.00.","severity":"breaking","affected_versions":">=25.10.00"},{"fix":"Ensure your system has CUDA Toolkit 12.x installed and configured correctly. If you need CUDA 11.x, install `pylibcudf-cu11` instead.","message":"The `pylibcudf-cu12` package is specifically compiled for CUDA Toolkit 12.x. Installing this package on a system with CUDA 11.x (or no CUDA toolkit) will likely result in `ImportError` or runtime failures due to ABI incompatibility.","severity":"gotcha","affected_versions":"All versions of `pylibcudf-cu12`"},{"fix":"Upgrade your CUDA Toolkit to version 12.x and install the corresponding `pylibcudf-cu12` package.","message":"In v25.08.00, `cudf` dropped support for CUDA 11 usages. This means that if you're upgrading from a significantly older `cudf` version and still using CUDA 11, you will encounter issues.","severity":"breaking","affected_versions":">=25.08.00"},{"fix":"If custom C++ code or extensions use this, they will need to be updated to use alternative resource management approaches, e.g., direct `rmm` usage.","message":"The C++ function `cudf::get_current_device_resource` was removed in v26.02.00. While a C++ change, it could affect custom extensions or deep integrations relying on this.","severity":"breaking","affected_versions":">=26.02.00"}],"env_vars":null,"search_vec":"'12':3 '26.4.0':48 'acceler':17 'align':55 'bind':10 'cadenc':54 'compon':59 'cu12':6 'cuda':2,63 'cudf':43 'data':33,69 'data-sci':68 'datafram':18,61,64 'ecosystem':27 'follow':50 'gpu':16,60 'gpu-acceler':15 'high':31,66 'high-perform':30,65 'higher':41 'higher-level':40 'layer':11 'level':42 'libcudf':13 'librari':19,44 'manipul':34 'month':52 'nvidia':25 'part':22 'perform':32,67 'primarili':36 'primit':35 'provid':29 'pylibcudf':1,5 'pylibcudf-cu12':4 'python':9 'rapid':26,58,62 'releas':53 'scienc':70 'use':37 'version':47","created_at":"2026-04-16T17:02:28.622636+00:00","updated_at":"2026-04-16T17:02:28.622636+00:00","problems":{"verify_error":"install timed out after 120s"},"ecosystem":"pypi","meta_description":null,"install_score":null,"quickstart_score":null,"quickstart_tag":null,"pypi_latest":"26.6.0","cli_name":"","cli_version":null,"type":"library","homepage":"https://rapids.ai","github":"https://github.com/rapidsai/cudf","docs":"https://docs.rapids.ai/api/cudf/stable/","changelog":null,"pypi":"https://pypi.org/project/pylibcudf-cu12/","npm":null,"openapi_spec":null,"status_page":null,"smithery":null,"categories":["data","ai-ml"],"base_url":null,"auth_type":null,"provenance":{"verified_status":"timeout","verified_at":"2026-06-28","last_verified":"2026-06-28","next_check":"2026-07-05","install_tag":null}}