Chinese AI chipmaker Biren Technology has announced a significant increase in the performance of its GPUs used for training large language models.

Biren has nearly doubled its capabilities through an upgrade developed with Infinigence AI, as announced on Tuesday.

Biren and Infinigence AI worked together in developing large language models by integrating Biren’s GPU cluster into Infinigence’s cloud platform, Infini AI.

Launched in April, Infini AI provides on-demand computing with chips from Nvidia, AMD, and now includes Biren’s software tools.

This cooperation arises as the US continues to limit China’s access to advanced semiconductor technology.