We’re excited to announce API support for Stable Audio 2.5 in ComfyUI — the first audio generation model designed specifically for enterprise-grade sound production at scale.
Stable Audio 2.5 introduces advancements in quality, speed, and control that make it ideal for commercial and professional use:
Fast generation: Create up to 3-minute tracks in under 2 seconds.
Smarter compositions: Richer multi-part structures with intros, developments, and outros.
Audio inpainting: Extend or remix audio by uploading your own clips and letting the model continue seamlessly.
Commercial safety: Trained on a fully licensed dataset for professional use.
Custom sound is an underutilized creative tool — but with Stable Audio 2.5 in ComfyUI, teams can embed distinct audio directly into their workflows, whether for ads, games, film, or immersive brand experiences.
Get Started
You can try Stable Audio 2.5 right now:
Update ComfyUI to the latest version
Search for “Stability AI audio” and you will find three related API nodes.
You can also find the Stable Audio 2.5 workflow in the templates.
Stable Audio 2.5 is built for dynamic and flexible results. Some things you can try today:
Generate a full-length ambient soundtrack with multiple evolving sections.
Create on-brand audio stingers or product sounds to reinforce a brand identity.
Extend an unfinished demo track by uploading it and letting the model inpaint a natural continuation.
Experiment with mood-driven prompts like “uplifting orchestral score with lush synthesizers” or “dark cinematic tension with sparse percussion”.
More Info
Stable Audio 2.5 represents a new step in multimodal creation workflows. By combining audio with image, video, and text pipelines inside ComfyUI, creators can build truly integrated experiences.
To learn more:
Visit our docs for more details.
Visit StableAudio.com.
Follow us on X, LinkedIn, and join the Comfy Discord to share your results.