Baichuan-7B vs LLMs-from-scratch
A neutral, constraint-first comparison - live GitHub stats and typed relationships, not marketing.
| Baichuan-7B | LLMs-from-scratch | |
|---|---|---|
| Tagline | A large-scale pretraining language model. | Implement a ChatGPT-like LLM in PyTorch from scratch |
| Stars | 5.7k | 99k |
| Forks | 501 | 15k |
| Open issues | 88 | 4 |
| Language | Python | Jupyter Notebook |
| License | Apache-2.0 | Other |
| Last pushed | Jul 18, 2024 | Jun 2, 2026 |
| Categories | LLM Frameworks | Model Training, LLM Frameworks |
Baichuan-7B
Baichuan-7B is a large-scale pretrained language model developed by BaiChuan-Inc, based on the Transformer structure. It features 7 billion parameters and has been trained on approximately 1.2 trillion tokens, supporting bilingual Chinese and English languages with a context window length of 4096 tokens.
Python
LLMs-from-scratch
Repository containing code and instructions for developing, pretraining, and finetuning a GPT-like large language model (LLM) using PyTorch.
Jupyter Notebook