KIDBench - Cultural Alignment (Nigeria): leaderboard
KIDBench Cultural Alignment Nigeria rates whether LLM answers to children aged 7 to 11 adapt guidance, help-seeking pathways and social expectations to Nigeria, when the system instruction names the child age and country (KIDBench, 2026). A DeepSeek-V4-Pro judge with country-specific rules scores each answer from 1 to 5, averaged over ten categories; higher is better.
Metric: Cultural Alignment score (1-5): whether the response adapts guidance, help-seeking pathways and social expectations to Nigeria, rated by a DeepSeek-V4-Pro judge with country-specific cultural rules (temperature 0), on the 500 implicit-cue KIDBench prompts in English with a system instruction naming a child aged 7 to 11 in Nigeria, mean over the 10 categories, single-turn; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.6 27B | 4.88 | #351 |
| 2 | Gemini 3.1 Flash Lite | 4.69 | #371 |
| 3 | Gemma 4 31B (IT) | 4.63 | #350 |
| 4 | Gemma 3 12B | 4.62 | #2059 |
| 5 | DeepSeek V4 Flash | 4.59 | #267 |
| 6 | GPT-5 Mini | 4.55 | #364 |
| 7 | Claude Haiku 4.5 | 4.47 | #583 |
| 8 | Qwen 3.5 4B | 4.06 | #1241 |
| 9 | Gemma 3 4B | 4.01 | #5339 |
| 10 | Llama 3.3 70B Instruct | 3.92 | #1385 |
| 11 | Qwen 3 8B | 3.91 | #1873 |
| 12 | Llama 3.1 8B Instruct | 3.49 | #4600 |
| 13 | KIDBench Llama-3.2-3B (checkpoint unspecified) | 3.39 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=kidbench-cultural-alignment-nigeria · How It Works · Data refreshed daily, snapshot 2026-11-21.