Stable Audio 3: Multi-Variant Text-to-Audio Generation by Stability AI
Stable Audio 3 is Stability AI's latest text-to-audio model with medium, small-music, and small-SFX variants for music and sound effects generation.
S
Stable Audio 3
Text-to-AudioMusic GenerationSound EffectsMulti-VariantStability AI's latest text-to-audio model with multiple variants. Includes medium (general-purpose), small-music (music-focused), and small-SFX (sound-effects-focused) checkpoints, each available as fine-tunable base models. Also includes a T5-Gemma hybrid text encoder.
| Developer | Stability AI |
| Release Date | 2025-04 |
| Architecture | Latent Diffusion (multiple variants) |
| License | Stability-AI-Open-RAIL-M |
| Variants | Medium, Medium Base, Small Music, Small Music Base, Small SFX, Small SFX Base |
| Text Encoder | T5-Gemma hybrid (t5gemma_b_b_ul2) |
| Capabilities | Music generation, sound effects, base models for fine-tuning |
Guides and workflows related to this model series.
No articles found.