KIDBench (No Cue) - Moral Guidance: leaderboard

KIDBench No Cue Moral Guidance rates whether an LLM answer to a child aged 7 to 11 encourages prosocial choices, responsible help-seeking and respect for others, when the prompts give no sign that the user is a child (KIDBench, 2026). A DeepSeek-V4-Pro judge scores each answer from 1 to 5, averaged over ten safety-relevant categories; higher is better.

Metric: Moral Guidance score (1-5): whether the response encourages prosocial choices, responsible help-seeking and respect for others, 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, no-cue prompts (no sign that the user is a child); higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1DeepSeek V4 Flash3.47#267
2Qwen 3.6 27B3.36#351
3Llama 3.3 70B Instruct3.3#1385
4GPT-5 Mini3.28#364
5Qwen 3 8B3.18#1873
6Qwen 3.5 4B3.15#1241
7Claude Haiku 4.53.11#583
8Gemma 3 12B3.05#2059
9Gemini 3.1 Flash Lite3.02#371
10Gemma 4 31B (IT)2.9#350
11Llama 3.1 8B Instruct2.86#4600
12Gemma 3 4B2.85#5339
13KIDBench Llama-3.2-3B (checkpoint unspecified)2.84

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

Interactive version: theaggregate.ai/benchmark?slug=kidbench-no-cue-moral-guidance · How It Works · Data refreshed daily, snapshot 2026-11-21.