Agent Arena ranks Tencent's Hy4 #14 overall, at about 62% less cost per task than Kimi K3.1 Pro

Arena says Tencent's Hy4 landed at #14 overall on its Agent Arena, which scores models on real agentic sessions from its user community. Arena says the result draws on more than 8,500 long-horizon sessions and a net improvement of 4.7%, and that Hy4 joins the Agent Arena Pareto frontier of score against cost.

According to Arena:

• Cost: Hy4 costs $0.55 per task against $1.46 for Kimi K3.1 Pro (Max), about 62% less, with scores within 1 percentage point of each other.
• Open models: Hy4 is proprietary for now; once its weights are released, Arena says it would tie Kimi K3.1 Pro for #1 among open models.
• Against the preview: this is 3.6 percentage points above the Hy4 preview, which ranked #25 with +1.1%.
• Signals: Confirmed Success #11 (+8.7%, up from #16); Praise vs Complaint #18 (+3.8%, up from #24); Steerability #14 (+5.7%, up from #27); Bash Reliability #20 (+6.2%, up from #41).

These are Arena's figures, drawn from its own users' sessions.