Phun-Bench - Rhyming Sentence Generation (1-Rhyme): leaderboard

Metric: Exact-match rate (%) of the rhyme group of the generated sentence's last character with the reference lyric line (1,000 lines; copying the reference ending disqualifies); thinking models run on 300 sampled items per setting, DeepSeek-V3 and GPT-4o on 500, the others on the full set; recommended sampling temperatures; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.

Top models

#ModelScore
1DeepSeek R1 052897
2GPT-4o (2024-11-20)82.7
3DeepSeek V3 (0324)81.6
4Qwen 3 8B (Non-reasoning)39.9
5GLM-4 9B Chat16
6Llama 3.2 3B Instruct11.1

Interactive version: theaggregate.ai/benchmark?slug=phun-bench-rhyming-sentence-generation-1-rhyme · How It Works · Data refreshed daily, snapshot 2026-09-29.