YapBench: leaderboard
Measures LLM verbosity on brevity-ideal prompts. Models are scored on excess length beyond a minimal baseline answer using YapScore and YapIndex metrics.
Metric: YapIndex (lower is better). Source: huggingface.co. Saturation forecast: Not forecast. 108 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol | 18.5 | #19 |
| 2 | GPT-3.5 Turbo | 22.7 | #865 |
| 3 | GPT-5.6 Luna | 27.8 | #68 |
| 4 | GPT-5.4 | 40.7 | #79 |
| 5 | Kimi K2 0905 | 44.7 | #288 |
| 6 | GPT-4 | 51.2 | #438 |
| 7 | GPT-5.3 Codex | 64.8 | #69 |
| 8 | GLM-4.7 Flash | 65 | #509 |
| 9 | GLM-4.7 | 68.7 | #192 |
| 10 | Grok 4 Fast | 70.2 | #254 |
| 11 | Kimi K2.7 Code | 72 | #89 |
| 12 | GPT-4.1 Mini | 75.7 | #355 |
| 13 | GLM-5.2 | 76.5 | #73 |
| 14 | Grok 4.5 | 83.8 | #48 |
| 15 | GPT-5.1 Codex | 85.3 | #159 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=yapbench · How It Works · Data refreshed daily, snapshot 2026-10-20.