I remember when people were saying 'It's useless to open-source big models because nobody will be able to run them fast'....

I remember when people were saying "It's useless to open-source big models because nobody will be able to run them fast".... Cerebras is now running Kimi K2.6 – a trillion parameter model – in enterprise trials. At ~1,000 tokens/s, this is the fastest frontier model performance ever measured by Artificial Analysis @ArtificialAnlys.

Source: Clem Delangue (X) | 2026-05-19