Llama-Chinese curates and presents Chinese-language resources and models related to Llama family, including Atom/Llama4/Llama3/Llama2 models, with a focus on Chinese pretraining, fine-tuning, and deployment resources. The repo lists numerous model references, data sources, and deployment guidance in its README.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
Llama中文社区 appears to be a repository and project hub that aggregates Chinese-language resources for Llama-based models, including open-source pre-trained models (Atom, Llama-4, Llama-3, Llama-2 variants) and related tooling and deployment guides. The README provides model tables and download links for multiple categories (pre-training, chat, multi-modal variants) and mentions community resources and latest updates.
How it works
The repository functions as a centralized index of models and related resources, offering direct links to model hosting locations (e.g., HuggingFace, ModelScope, WiseModel) and documenting available variants (e.g., Atom-7B, Llama-4-Scout, Llama-3.2-1B, Llama2-13B-Chat). It also outlines data sources and optimization notes for Chinese pretraining, and lists deployment and API options through external platforms.
Getting started
- The README includes sections with model download references and usage recommendations. Install/usage commands are provided in the sections describing how to use various models and platforms; however, the exact shell commands to install or run aren’t included in the truncated excerpt provided here. If needed, consult the linked model pages (HuggingFace/ModelScope) for exact commands.
Recent releases
- The Releases section shows: latest 0: none.
Traction
- Stars: 14748
- Forks: 1297
- Open issues: 195
Behind the repo
- The repository highlights collaborations with HuggingFace, ModelScope, and WiseModel through listed model entries and references in the README.
Caveats
- License is not listed.
- Created: 2023-07-19
- Last push: 2025-04-06