Skip to content

Chinese Tiny LLM

A 2B parameter model trained mainly on Chinese

Visit official site
Category AI models

What Chinese Tiny LLM is

The model is described as a Chinese focused large language model with 2 billion parameters, trained from scratch on a large Chinese corpus.

What you can do with it

  • Work on Chinese language tasks
  • Handle English text and programming code
  • Study the released training resources

What to watch out for

  • The stated corpus is 1.2 trillion Chinese tokens
  • CHC-Bench is the named benchmark
  • Intermediate checkpoints are released

Pros & cons

✓ What we like

  • English and programming code are handled
  • The data filtering process is opened

! What to watch out for

  • Pretraining mainly uses Chinese data
  • Training starts from scratch

FAQ

How large is the model?

The material states 2 billion parameters.

What is opened to researchers?

The material lists data filtering, training dynamics, training and evaluation data and intermediate checkpoints.

Last reviewed: 2026-11-15

More AI education tools

View all →

How we review