DeepZang, World's First Tibetan Large Language Model, Launched - Tibetan AI Era Begins

For many Tibetan speakers, interacting with AI in their native language has long been an unmet need. Recently, DeepZang, the world's first Tibetan large language model, was officially launched in Lhasa, marking the start of the Tibetan AI era and filling a global technological gap in this field.

Developed independently by Tibet Jueluo Digital Industry Management Co., Ltd., DeepZang is China's first Tibetan language model to complete the national generative AI algorithm and model filing. Its app supports intelligent interaction in Tibetan, Chinese, and English, integrating dialogue, real-time translation, and speech transcription functions. It has accumulated nearly 70 million national standard Tibetan parallel corpora and over 30,500 hours of voice data for the three major Tibetan dialects.

The model also received a certificate from the World Record Certification Agency as the 'World's First Tibetan Large Language Model', confirming its global pioneering status and helping to enhance China's voice in global multilingual AI governance.