If Jev Router were on the Agent Arena leaderboard, it would post a +3.6% net improvement at a median cost of $0.155 per task.
It would rank below DeepSeek V4.1 Flash, the model it routes to most, but beat it on Steerability (+10.0% vs -0.2%) and Praise vs Complaint (+6.3% vs +2.1%).
Where DeepSeek does better is on Confirmed Success (+8.3% vs +2.2%), Bash Recovery (+7.8% vs -0.7%) and Tool Hallucination (+0.4% vs +0.2%).
The key signal here is steerability - it demonstrates Jev’s strong capability to respond to user corrections. In fact, the router’s score is not far from that of Claude Opus 5.5 (High) - only 0.48 percentage points lower.
