wenge-research/YAYI2
YAYI 2 is a 30B parameter open-source multilingual large language model developed by Wenge Research.
Not currently ranked — collecting fresh signals.
star history
YAYI 2 is a 30B parameter open-source large language model developed by Wenge Research (中科闻歌). It is trained on over 2 trillion high-quality multilingual tokens using a Transformer architecture. The project includes Base and Chat versions, with support for fine-tuning (full-parameter and LoRA), inference code, and technical documentation.
Frequently asked
- What is wenge-research/YAYI2?
- YAYI 2 is a 30B parameter open-source multilingual large language model developed by Wenge Research.
- Is YAYI2 open source?
- Yes — wenge-research/YAYI2 is open source, released under the Apache-2.0 license.
- What language is YAYI2 written in?
- wenge-research/YAYI2 is primarily written in Python.
- How popular is YAYI2?
- wenge-research/YAYI2 has 2.8k stars on GitHub.
- Where can I find YAYI2?
- wenge-research/YAYI2 is on GitHub at https://github.com/wenge-research/YAYI2.