DailySafe (Single-Turn Consolidated) - Helpfulness Rate (HR): leaderboard

DailySafe (arXiv 2611.26298v1) dynamic multimodal safety benchmark (Helpfulness Rate (HR)) over 195 everyday multi-turn image-text cases.

Metric: Single-turn consolidated Helpfulness Rate (%, 3-run mean). Source: arxiv.org. Saturation forecast: Around February 2027. 16 models tracked. Updated 22 Nov 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 5.595.21#9
2GPT-6 Sol91.62#33
3GPT-5.6 Sol90.43#36
4Gemini 3.8 Flash89.57#43
5Gemini 3.1 Pro (Preview)82.91#102
6Gemini 3.5 Flash81.71#110
7Kimi K375.21#62
8GLM-5.3 Flash70.09#137
9Claude Sonnet 560.51#124
10Qwen3.8-Flash-Next59.66#122
11Qwen 3.8 27B45.13#232
12MiMo-V2.6-Pro40.17#94
13DeepSeek V4.1 Flash34.7#69
14MiMo-V2.6-Flash24.62#183
15Qwen 3 VL 8B (Thinking)6.32#924

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=dailysafe-single-turn-consolidated-helpfulness-rate-hr · How It Works · Data refreshed daily, snapshot 2026-11-22.