KIDBench (Implicit Cue) - Content Safety: leaderboard

KIDBench Implicit Cue Content Safety rates whether an LLM answer to a child aged 7 to 11 avoids harmful, explicit or misleading content, when the prompt wording or situation suggests a child speaker without stating an age (KIDBench, 2026). A DeepSeek-V4-Pro judge scores each answer from 1 to 5, averaged over ten safety-relevant categories; higher is better.

Metric: Content Safety score (1-5): whether the response avoids harmful, explicit or misleading content, rated 1-5 by a DeepSeek-V4-Pro judge with the KIDBench rubric and category-specific rules, temperature 0 for responder and judge, averaged on the KIDBench child-safety prompts for children aged 7 to 11 (50 per category, 10 categories), single-turn, implicit-cue prompts (wording or situation suggests a child speaker, age not stated); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 3.6 27B4.68#351
2Gemini 3.1 Flash Lite4.57#371
3DeepSeek V4 Flash4.55#267
4Gemma 4 31B (IT)4.49#350
5Claude Haiku 4.54.37#583
6GPT-5 Mini4.35#364
7Qwen 3.5 4B4.31#1241
8Gemma 3 12B4.28#2059
9Llama 3.3 70B Instruct4.25#1385
10Qwen 3 8B3.99#1873
11Llama 3.1 8B Instruct3.86#4600
12Gemma 3 4B3.79#5339
13KIDBench Llama-3.2-3B (checkpoint unspecified)3.57

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=kidbench-implicit-cue-content-safety · How It Works · Data refreshed daily, snapshot 2026-11-21.