A phonebook for Mandarin-speaking machines

What it does

This is a curated awesome-list that catalogs open-source libraries, datasets, and commercial services for Chinese natural language processing. It groups hundreds of resources—from Tsinghua’s THULAC and Baidu’s LAC to Jieba and Chinese-CLIP—into categories like word segmentation, information extraction, chatbots, and multi-modal retrieval. Think of it as a card index for a field where tool names are often in Mandarin and documentation is scattered across universities and tech giants.

The interesting bit

The list deliberately mixes academic heavyweights with lightweight Python one-offs, so you can compare a Java toolkit from the Chinese Academy of Sciences side-by-side with a modern neural pipeline. It also tracks non-code resources: corpora, industry vendors, and learning materials, making it closer to a field guide than a simple package index.

Key highlights

• Covers the full pipeline: tokenization, NER, relation extraction, text classification, dialogue systems, and CLIP-style vision-language models.
• Surfaces institutional tools often buried in Chinese university repos: THULAC, LTP, FudanNLP, and pkuseg.
• Includes a dedicated section on Chinese-language corpora and knowledge graphs such as OpenKG and CN-Probase.
• Lists commercial APIs and academic organizations alongside open-source code.
• Maintained in both English and Chinese, which matches the bilingual reality of the ecosystem.

Verdict

Worth bookmarking if you build anything that processes Chinese text and need to survey the landscape before picking a dependency. Skip it if you are looking for a single framework with a unified API; this is pure curation, not code.