CuPyNumeric Parallel Data Load

CuPyNumeric Parallel Data Load is a Data skill for Claude Code, published by NVIDIA in skills.

3K stars345 forkson NVIDIA/skillsAdded 2026/08/17+3% in starsRepository updated 2026/08/17

agent-skillsagentic-aiai-agentsclaude-codecodexcoding-agentscudadeveloper-toolsllmomniversephysical-airagroboticssimulation

Install in seconds

Install CuPyNumeric Parallel Data Load

Copy CuPyNumeric Parallel Data Load into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/NVIDIA/skills/tree/main/skills/cupynumeric-parallel-data-load ~/.claude/skills/cupynumeric-parallel-data-load

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/NVIDIA/skills.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/cupynumeric-parallel-data-load/SKILL.md in NVIDIA/skills
Installs to: ~/.claude/skills/cupynumeric-parallel-data-load
Collection: One of 25 skills cataloged from this repository
Category: Data — 668 skills

What CuPyNumeric Parallel Data Load does

CuPyNumeric Parallel Data Load loads sharded, on-disk datasets (.npy, Parquet/Arrow, raw binary, HDF5) into a distributed cuPyNumeric ndarray using manual partition and leaf task launch for CPU, OMP, and GPU variants. Use it when no single-call loader fits, including when per-shard row counts differ across files.

CuPyNumeric Parallel Data Load is cataloged under Data on DirSkills. CuPyNumeric Parallel Data Load comes from a repository tagged agent-skills, agentic-ai, ai-agents, claude-code and codex.

Documentation

README

Excerpt

Parallel sharded data -> cupynumeric load

Why this skill exists. cupynumeric mirrors NumPy's array API, including cupynumeric.load for a single .npy file. Beyond that, file loading lives in Legate, not cupynumeric:

Format	Built-in loader
Single `.npy`	`cupynumeric.load(path)` (NumPy-API parity)
HDF5 (single file)	`legate.io.hdf5.from_file` / `from_file_batched`
Sharded multi-file (any format), Parquet/Arrow, raw binary, custom layouts	No built-in loader — this skill.

This is the opening of the README. Read the full README on GitHub.

Frequently asked about CuPyNumeric Parallel Data Load

What else does NVIDIA publish alongside CuPyNumeric Parallel Data Load?
CuPyNumeric Parallel Data Load is one of 25 skills that DirSkills catalogs from NVIDIA/skills, the repository it ships in. Its siblings there include AIQ Deploy, AIQ Research and AMC RTSP Calibration. Each one is a separate skill with its own page in this directory, installs the same way CuPyNumeric Parallel Data Load does, and is maintained by NVIDIA in that same repository. The rest of the collection is listed on the NVIDIA/skills page.
How does CuPyNumeric Parallel Data Load compare to other Data skills?
CuPyNumeric Parallel Data Load ranks #249 by stars among the 668 Data skills in this catalog. The most-starred ones next to it are Benchmark Methodology, Jupyter Notebook and Solana. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of CuPyNumeric Parallel Data Load against them. Open each page to compare what they document and how they install.