Jev launches: a System One model for decisions
TypeSafe founder Diogo Almeida introduces Jev, its RLCD training method, and its focus on software decisions. The post includes a 2:56 launch video.

JevBench v1.4.1 adds six new systems with the top five unchanged. Best newcomer is Jev-Omni, a multimodal Gemma-based model, at #7 (51.34). Jev 1.13.0 still holds #1 with 63.29.
JevBench v1.4.1 is out. Six systems added, top five unchanged. Best newcomer: Jev-Omni, a multimodal Gemma-based model, at #7 (51.34). Jev 1.13.0 still holds #1 at 63.29. https://benchmarkheaven.com/jev-models