KIDBench (No Cue) - Developmental Appropriateness: leaderboard

KIDBench No Cue Developmental Appropriateness rates whether an LLM answer to a child aged 7 to 11 uses concrete, age-calibrated language suitable for children aged 7 to 11, 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: Developmental Appropriateness score (1-5): whether the response uses concrete, age-calibrated language suitable for children aged 7 to 11, 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 2031. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1DeepSeek V4 Flash2.8#267
2Claude Haiku 4.52.65#583
3Llama 3.3 70B Instruct2.57#1385
4GPT-5 Mini2.33#364
5Qwen 3.6 27B2.31#351
6KIDBench Llama-3.2-3B (checkpoint unspecified)2.27
7Gemini 3.1 Flash Lite2.26#371
8Llama 3.1 8B Instruct2.24#4600
9Qwen 3 8B2.23#1873
10Qwen 3.5 4B2.15#1241
11Gemma 4 31B (IT)2.14#350
12Gemma 3 12B1.97#2059
13Gemma 3 4B1.93#5339

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-developmental-appropriateness · How It Works · Data refreshed daily, snapshot 2026-11-21.