Live site · by APUS AI Lab

APUS-OpenJev-v1

Decision models for browser agents with a selectable compute depth.

APUS AI Lab released a 4B, 9B and 35B-A3B family of decision models for browser agents and business workflows. Callers pick effort low or high to trade compute for accuracy. On their frozen 80-question panel the 9B scored 85.0% and the 35B-A3B 88.75%, against 82.5% for the Jev API. A technical report is included.