FAD (lower is better)

We haven't written a description for this benchmark yet; see the card notes below for how each maker ran it.

Your workloadRAG over docs · 12K in · 600 out · 50K requests/mo · 30% cached

ModelScoreYour cost / mo
1Stable Audio 3.0 Medium · Stability AI
0.369 FADself-host
2Stable Audio 3.0 · Stability AI
0.358 FADself-host
3Stable Audio 2.5 · Stability AI
0.128 FADself-host
4Stable Audio 3.0 Medium · Stability AI
0.116 FADself-host
5Stable Audio 3.0 Medium · Stability AI
0.107 FADself-host
6Stable Audio 2.5 · Stability AI
0.106 FADself-host
7Stable Audio 3.0 · Stability AI
0.101 FADself-host
8Stable Audio 3.0 · Stability AI
0.1 FADself-host

“Best value”: no other model here reports a higher score at a lower cost for your workload. Scores are self-reported by each maker with different settings, so treat close calls as ties.

How each maker reported it

Stable Audio 3.0 Medium paper, Tables 3–5
Stability's own results. FAD (lower is better) and CLAP (higher is better) use LAION-CLAP embeddings. OVL, REL and MUS are mean opinion scores from 14 listeners. Time is wall-clock seconds on an H200. Stable Audio 2.5 was evaluated on music only; Stable Audio Open on sound effects only. The Small column is small-music for music rows and small-sfx for sound-effect rows.
Stable Audio 3.0 paper, Tables 3–5
Stability's own results. FAD (lower is better) and CLAP (higher is better) use LAION-CLAP embeddings. OVL (overall quality), REL (text relevance) and MUS (musicality) are mean opinion scores from 14 listeners. Time is wall-clock seconds on an H200. Stable Audio 2.5 was evaluated on music only.
Stable Audio 2.5 Stable Audio 3 paper (arXiv 2605.17991), Tables 3–4
The 2.5 launch post has no benchmark numbers. These are Stability's own later comparisons in the Stable Audio 3 paper. FAD lower is better; CLAP higher is better; OVL, REL and MUS are mean opinion scores from 14 listeners; time is seconds on an H200.
Stable Audio 3.0 Medium paper, Tables 3–5
Stability's own results. FAD (lower is better) and CLAP (higher is better) use LAION-CLAP embeddings. OVL, REL and MUS are mean opinion scores from 14 listeners. Time is wall-clock seconds on an H200. Stable Audio 2.5 was evaluated on music only; Stable Audio Open on sound effects only. The Small column is small-music for music rows and small-sfx for sound-effect rows.
Stable Audio 3.0 Medium paper, Tables 3–5
Stability's own results. FAD (lower is better) and CLAP (higher is better) use LAION-CLAP embeddings. OVL, REL and MUS are mean opinion scores from 14 listeners. Time is wall-clock seconds on an H200. Stable Audio 2.5 was evaluated on music only; Stable Audio Open on sound effects only. The Small column is small-music for music rows and small-sfx for sound-effect rows.
Stable Audio 2.5 Stable Audio 3 paper (arXiv 2605.17991), Tables 3–4
The 2.5 launch post has no benchmark numbers. These are Stability's own later comparisons in the Stable Audio 3 paper. FAD lower is better; CLAP higher is better; OVL, REL and MUS are mean opinion scores from 14 listeners; time is seconds on an H200.
Stable Audio 3.0 paper, Tables 3–5
Stability's own results. FAD (lower is better) and CLAP (higher is better) use LAION-CLAP embeddings. OVL (overall quality), REL (text relevance) and MUS (musicality) are mean opinion scores from 14 listeners. Time is wall-clock seconds on an H200. Stable Audio 2.5 was evaluated on music only.
Stable Audio 3.0 paper, Tables 3–5
Stability's own results. FAD (lower is better) and CLAP (higher is better) use LAION-CLAP embeddings. OVL (overall quality), REL (text relevance) and MUS (musicality) are mean opinion scores from 14 listeners. Time is wall-clock seconds on an H200. Stable Audio 2.5 was evaluated on music only.

Only each model's own maker's number is shown; scores a maker reports for rival models are left out.