← All stories
Markets 3 sources

Meituan open sources LongCat-2.0, a 1.6 trillion parameter coding model trained entirely on Chinese chips

Covered by 3 sources · 3 articles

3 outlets - including Cryptopolitan, Defi-planet and Cryptobriefing - are covering this story. Meituan claims it trained the 1.6 trillion parameter model on domestic Chinese hardware, avoiding Nvidia GPUs altogether.

Covered by

All coverage

Cryptopolitan Jun 30, 23:01 UTC

Meituan drops 1.6-trillion-parameter LongCat-2.0 trained on Chinese chips

Meituan claims it trained the 1.6 trillion parameter model on domestic Chinese hardware, avoiding Nvidia GPUs altogether.

Meituan Announces AI Model Trained Entirely on Chinese-made Chips
DeFi Planet Jun 30, 13:01 UTC

Meituan Announces AI Model Trained Entirely on Chinese-made Chips

Chinese technology company Meituan has introduced LongCat-2.0, a new large language model that it says was trained entirely on Chinese-made AI chips. The company described it as the world’s first trillion-parameter AI system built and opera…

Crypto Briefing Jun 30, 05:44 UTC

Meituan open sources LongCat-2.0, a 1.6 trillion parameter coding model trained entirely on Chinese chips

Meituan's LongCat-2.0 challenges US chip export controls, potentially reshaping global AI dynamics and fostering Chinese tech independence. The post Meituan open sources LongCat-2.0, a 1.6 trillion parameter coding model trained entirely on…