RedVox: leaderboard
Metric: Unsafe response rate (%; share of responses a GPT-5.5 judge labels unsafe (endorsing, supporting or complying with the harmful or stereotyped request, whether or not the request was understood); all five languages and request types of the 3,414 released entries). Source: arxiv.org. Saturation forecast: Estimated already saturated. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Flash Lite | 2.6 |
| 2 | Gemini 3.1 Pro (Preview) | 3.1 |
| 3 | Qwen3 Omni 30B A3B Instruct | 3.4 |
| 4 | Qwen2-Audio-7B-Instruct | 10.9 |
| 5 | Phi-4 Multimodal Instruct | 16.1 |
| 6 | Voxtral-Small-24B-2507 | 21.9 |
Interactive version: theaggregate.ai/benchmark?slug=redvox · How It Works · Data refreshed daily, snapshot 2026-09-26.