DeepSWE
Patronus AI Details In-House Post-Training Stack and First Results on Qwen3.8-27B
Patronus AI has published a writeup of the post-training stack it built in-house, along with first results from training the open-weight model Qwen3.8-27B. According to…
Replit lets its agent's model pick subagents, effort and delegation itself
Replit argues that model routers are limited by design, and says its Replit Agent instead lets the main model decide which subagents to use, how large they are and how…