Tencent/PatrickStar

PatrickStar enables Larger, Faster, Greener Pretrained Models for NLP and democratizes AI for everyone.

48
/ 100
Emerging

This project helps AI engineers and researchers train very large language models without needing enormous computing resources. It takes your existing PyTorch model code and a configuration, allowing you to train models significantly larger than typically possible on your available GPUs by efficiently using both GPU and CPU memory. This is for machine learning practitioners working with large-scale natural language processing.

779 stars.

Use this if you are encountering 'out-of-memory' errors when trying to train large language models and want to use fewer GPUs or train much bigger models with your current hardware.

Not ideal if you are working with small models or have abundant GPU resources and do not face memory constraints.

large-language-models NLP-training AI-resource-optimization deep-learning-scalability pre-trained-models
No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

779

Forks

59

Language

Python

License

BSD-3-Clause

Last pushed

Nov 18, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/Tencent/PatrickStar"

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