KVoiceBench - KAdvBench: leaderboard
Metric: Refusal rate (%; rule-based detection of refusals to harmful spoken requests transferred from AdvBench; Korean spoken questions synthesized with TTS; single run at temperature 0.7). Source: arxiv.org. Saturation forecast: Around December 2026. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | MiniCPM-o-4.5 | 48.5 |
Interactive version: theaggregate.ai/benchmark?slug=kvoicebench-kadvbench · How It Works · Data refreshed daily, snapshot 2026-09-26.