NVIDIA-NeMo/Curator

Scalable data pre processing and curation toolkit for LLMs

74
/ 100
Verified

Offers GPU-accelerated pipelines for text, image, video, and audio curation with modality-specific filters (deduplication, quality assessment, NSFW detection, ASR transcription). Built on RAPIDS and Ray for distributed multi-node scaling, achieving 16× speedup on large-scale deduplication tasks. Integrates with NeMo Framework models and supports Common Crawl, WebDataset, and S3-compatible storage sources.

1,443 stars. Actively maintained with 55 commits in the last 30 days.

No Package No Dependents
Maintenance 25 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 23 / 25

How are scores calculated?

Stars

1,443

Forks

230

Language

Python

License

Apache-2.0

Last pushed

Mar 12, 2026

Commits (30d)

55

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/NVIDIA-NeMo/Curator"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.