SAVRN
Search Contact SAVRN

Organization

Hugging Face Code Research

HuggingFaceCode

Exploring code models and datasets

Models in Library0
Datasets in Library1
Models on Hugging Face
Followers175

Datasets

Dataset · Text generation

stack-v3-train

Hugging Face Code Research

The Stack v3 is the largest, most up-to-date open dataset of source code, crawled directly from GitHub and built to pre-train code LLMs with full-repository context. It is the successor to The Stack v2 and, like its predecessor, is released to make the training of code models more open, reproducible, and transparent. If your code is included in this version and you would like it removed for any reason, please use the "Am I in The Stack?" Space to submit a removal request. Repositories that opted out are removed from the dataset before each patch release. Two things set v3 apart from earlier releases of The Stack: 1. The file contents are included inline. The decoded UTF-8 source text is…

Publicly accessible odc-by 100M<n<1B