Guan Wang's Sapient Says It Trained a 1B-Parameter Model for About $1,500

Guan Wang, CEO of Sapient Intelligence, claims his researchers trained a 1 billion parameter language model from scratch for approximately $1,500, according to a June 10 VentureBeat report. The model, called HRM Text, is a language adaptation of Sapient's Hierarchical Recurrent Model architecture first introduced in 2025. The claim challenges the prevailing economics of AI development by suggesting that large-scale models can be built at a fraction of typical costs.

Guan Wang, CEO of Sapient Intelligence, is putting a number on a bet that runs against the dominant economics of AI: his researchers say they trained a 1 billion parameter language model from scratch for about $1,500, according to a VentureBeat report published June 10. The claim centers on HRM Text, Sapient Intelligence's language adaptation of a Hierarchical Recurrent Model HRM architecture that Sapient first introduced in 2025, according to VentureBeat. Wang's argument is not that enterp...