Artificial Analysis launches AA-Music v1.1 benchmarks with Suno v6 leading

Artificial Analysis launched AA-Music-Vocal v1.1 and AA-Music-Instrumental v1.1, the updated evaluation benchmarks powering its Music Arena. The benchmarks use 1,000 new prompts across 17 genres, with outputs rated by a recruited panel of evaluators.

On the updated leaderboards, Suno v6 placed first on both benchmarks, leading Suno v6-mini by 26 Elo on Vocal and 31 Elo on Instrumental. Mureka V9.5 ranked third on Vocal, outpacing Mureka V9 by 51 Elo. On Instrumental, Mureka V9, Mureka V9.5, Lyria 3 Pro, and Lyria 3.5 are statistically tied between third and sixth place within a 5 Elo spread. MiniMax Music 3.0 led open-weights models at 11th on Vocal and 13th on Instrumental, with Stable Audio 3 Medium within 3 Elo of it on Instrumental.