#sapient Tech News.
1 story in the last 7 days
The latest sapient news, distilled by AI into sharp ~100-word summaries. ByteBrief tracks sapient across dozens of tech sources and brings you only what matters, updated hourly. Tap any story for the full brief, or open the original source.
AIVentureBeat2 days ago
Foundation model trained from scratch for $1,500Sapient researchers trained HRM-Text, a foundation LLM, from scratch for about $1,500. The model replaces standard Transformers with a Hierarchical Recurrent Model that decouples computation into strategic and execution layers. Training uses only instruction data instead of internet-scale text, dramatically cutting cost.
Summaries by ByteBrief