---
license: apache-2.0
task_categories:
- text-generation
pretty_name: Jian Dataset
size_categories:
- n>1T
language:
- en
- fr
- es
- it
- zh
- ar
- ru
- pt
tags:
- text
- code
- web-corpus
- multilingual
- large-scale
description: >-
  Jian is a trillion-token-scale multilingual corpus of web text and code
  designed for large language model pretraining.
---

Jian is a trillion-token-scale multilingual corpus of web text and code designed for large language model pretraining.