#1111th of 14 in Learning and cookbooks
happy-llm
Chinese-language course on LLM theory and building a LLaMA2-style model
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 34.4k
- License
- custom license
- Last commit
- Aug 2026
- Last release
- Jan 2026
- Language
- Jupyter Notebook
27,439 on 24 Mar 202634,355 on 13 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Happy-LLM is a Datawhale course in eight chapters, written in Chinese with an English README. It covers NLP basics, the Transformer and pretrained model families, then implements LLaMA2 in PyTorch, trains a tokenizer, and runs pretraining, SFT, LoRA/QLoRA, RAG, agents and GRPO-based agentic RL. Two 215M checkpoints from chapter 5 are published on ModelScope.
Who it is for: Students and engineers learning how LLMs are built and trained, reading Chinese
Strengths
- Covers Transformer basics through pretraining, SFT, LoRA/QLoRA, RAG, agents and agentic RL
- Publishes 215M base and SFT checkpoints plus a ModelScope studio to try them
- Free PDF and companion slides available from the releases page
- Dependencies split per chapter, with an environment preparation guide
Weaknesses
- Main content is in Chinese; only the README has an English version
- Content license is CC BY-NC-SA 4.0, which restricts commercial use
- Hardware requirements are not stated in the README, only in a separate doc
- Extra Chapter blog section is marked as in progress
What it needs
- Needs PyTorch, Transformers
- Models: LLaMA2, Happy-LLM-215M-base, Happy-LLM-215M-sft
Also in Learning and cookbooks
See all 14| Rank | Project | Score |
|---|---|---|
| 1 | ai-engineering-from-scratchHands-on curriculum for building AI systems from math to agents | 82 out of 100 |
| 2 | generative-ai-for-beginners21-lesson course on building generative AI apps in Python and TypeScript | 81 out of 100 |
| 3 | hello-agentsChinese-language course on building LLM agents from scratch | 76 out of 100 |
| #4 | llms-from-scratchCode for the book on building a GPT-style LLM in PyTorch | 75 out of 100 |
| #5 | ai-agents-for-beginners18-lesson course on building AI agents with Microsoft Agent Framework | 74 out of 100 |