Stable Audio Open Small β€” Core AI (on-device music generation)

The model zoo's first MUSIC / AUDIO generation model for Apple Core AI. Type a prompt, get ~11s of 44.1 kHz stereo audio β€” generated entirely on-device on Apple Silicon. A community port of stabilityai/stable-audio-open-small (Stability AI + Arm) to Core AI.

A latent diffusion text-to-audio model: a T5 text encoder conditions a DiT (diffusion transformer) that denoises a latent over 8 rectified-flow steps, then an Oobleck VAE decodes the latent to a waveform. Distilled (ARC) for few-step generation, so it's fast.

What's in the bundle (macos/)

Three Core AI .aimodel bundles + a tiny host sampler loop:

bundle role I/O
sa_cond_fp16b T5-base encoder + number conditioner input_ids[1,64], attention_mask[1,64], seconds_norm[1] β†’ cross_attn_cond[1,65,768], global_embed[1,768], cond_mask[1,65]
sa_dit_fp16 diffusion transformer (run 8Γ—) x[1,64,256], t[1], cross_attn_cond, global_embed, cross_attn_cond_mask β†’ v[1,64,256]
sa_vae_fp16 Oobleck VAE decoder latent[1,64,256] β†’ audio[1,2,524288]

Host loop (StableAudioRunner): tokenize (T5, t5_tokenizer/) β†’ conditioner β†’ start from Gaussian noise β†’ 8-step rectified-flow euler x = x + (t_next βˆ’ t)Β·v over the fixed schedule [1.0, .9944, .9845, .9579, .8909, .7455, .5125, .2739] β†’ 0 β†’ VAE decode β†’ 44.1 kHz stereo wav. No KV cache, no CFG (cfg_scale 1.0 β€” the model is ARC-distilled).

Performance (M4 Max, GPU)

metric value
8-step DiT ~200 ms (25 ms/step)
VAE decode ~185 ms
total 0.4 s for ~11.9 s of audio (30Γ— real-time)
size fp16, ~1.0 GB (DiT 651M + cond 210M + VAE 149M)

Numerics: each bundle engine-gated vs the reference at cos β‰₯ 0.9999; full pipeline reproduces the reference audio exactly.

Roadmap

  • iPhone (h18p) build β€” bundles AOT-compile; device RTF pending
  • int8 (further size cut)
  • a music-generation tab in the zoo app

Credits & license

A community Core AI conversion β€” all credit to Stability AI (and Arm) for Stable Audio Open Small; T5 text encoder by Google. This bundle is governed by the Stability AI Community License (free for non-commercial use and for commercial use under $1M annual revenue; review the license before use). No retraining β€” conversion only.

Part of the Core AI model zoo.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for mlboydaisuke/Stable-Audio-Open-Small-CoreAI

Finetuned
(3)
this model