YOMI-Bench - Rhyme Selection (Kanji): leaderboard

Metric: Accuracy (%; four-option choice of the word sharing the vowel sequence of a word written in kanji; 120 items; mean over five paraphrased prompts; accuracy as a fraction x100). Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.

Top models

#ModelScore
1GPT-5100
2Claude Sonnet 4.595.59
3Gemini 2.5 Flash93.8
4GPT-4o58.2
5Mistral Medium 3.134.6
6Ministral-8B-Instruct-241029.6
7Llama-3.1-Swallow-8B-Instruct-v0.529.6

Interactive version: theaggregate.ai/benchmark?slug=yomi-bench-rhyme-selection-kanji · How It Works · Data refreshed daily, snapshot 2026-09-29.