boogiebench: leaderboard
Elo-ranked LLM music-composition arena where text models write Strudel JavaScript compositions from music prompts and users vote on the generated tracks.
Metric: Elo Rating. Source: www.boogiebench.com. Status: saturation imminent. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 1679 |
| 2 | GPT-5.4 | 1664 |
| 3 | GPT-5.2 | 1621 |
| 4 | Claude Opus 4.6 | 1620 |
| 5 | Claude Opus 4.5 | 1597 |
| 6 | Claude Sonnet 4.5 | 1590 |
| 7 | GPT-5 | 1535 |
| 8 | Gemini 2.5 Pro | 1459 |
| 9 | DeepSeek V3.1 Terminus | 1435 |
| 10 | GLM-4.6 | 1426 |
| 11 | GPT-OSS-120B | 1397 |
| 12 | MiniMax-M2 | 1397 |
| 13 | Kimi K2 0905 | 1361 |
Interactive version: theaggregate.ai/benchmark?slug=boogiebench · How It Works · Data refreshed daily, snapshot 2026-09-05.