Moffett AI, a Shenzhen-based developer of AI inference chips, announced on November 3 that it had closed a Series C+ round of nearly RMB 1 billion (about $140 million). It comes roughly five months after a previous round of similar size.
The money will go toward upgrading the company's sparse-computing chips, research on next-generation processors and collaboration with research institutions and industry. Sparse computing cuts the number of calculations needed to run a model by exploiting the zeros and weak connections in its data and operations.
The company says it has already deployed inference clusters in four regions of China and works with telecom operators. Its next-generation SP200 accelerator card is due before the end of the year.