Stability AI Stable Audio 2.5 API Node ComfyUI Official workflow example

This article will introduce how to use Stability AI Stable Audio 2.5 Partner node’s text-to-audio, audio-to-audio and audio-inpainting capabilities in ComfyUI

The Stability AI Stable Audio 2.5 Partner node allows you to use Stability AI’s latest audio generation model to create high-quality music through text prompts, audio transformations, and audio inpainting capabilities.

Stable Audio 2.5 is designed for enterprise use, featuring improved musical structure, better prompt adherence, and the ability to generate minutes-long compositions in seconds. The model offers three main workflows: Text-to-Audio for generating music from descriptions, Audio-to-Audio for transforming existing audio into new compositions, and Audio Inpainting for completing or extending existing tracks.

Trained exclusively on licensed audio, Stable Audio 2.5 is commercially safe and perfect for advertisers, game studios, and content creators who need professional-quality audio generation with enterprise-grade reliability.

To use the API nodes, you need to ensure that you are logged in properly and using a permitted network environment. Please refer to the API Nodes Overview section of the documentation to understand the specific requirements for using the API nodes.

Portable or self deployed users

Make sure your ComfyUI is updated.

• Download ComfyUI
• Update Guide

Workflows in this guide can be found in the Workflow Templates. If you can’t find them in the template, your ComfyUI may be outdated. (Desktop version’s update will delay sometime)

If nodes are missing when loading a workflow, possible reasons:

1. You are not using the latest ComfyUI version (Nightly version)
2. Some nodes failed to import at startup

Desktop or Cloud users

• The Desktop is base on ComfyUI stable release, it will auto-update when there is a new Desktop stable release available.
• Cloud will update after ComfyUI stable release.

So, if you find any core node missing in this document, it might be because the new core nodes have not yet been released in the latest stable version. Please wait for the next stable release.

Text-to-Audio Workflow

For text-to-audio, you can generate audio through text prompts. You need to describe the music you want to generate.

Download JSON Workflow

1. Modify the text prompt. You should use keywords to describe the music you want to generate.
2. (Optional) Modify the duration parameter. It’s 190 by default.
3. Click the Run button or use the shortcut Ctrl(cmd) + Enter to execute the audio generation. The audio will be saved to the ComfyUI/output/audio directory.

Audio-to-Audio Workflow

Audio-to-audio is basically music re-sampling. You can use it to generate new music from a given piece of music, or you can just hum a melody, and then the model will generate new music based on the input audio.

Download JSON Workflow

Download Input Audio

1. In this workflow, we have provided two nodes for you to input the audio（at least 6 seconds） you want to edit:
   • 1.1 Record Audio node: You can use it to record any of your music ideas, such as a hummed melody. It should be at least 6 seconds.
   • 1.2 LoadAudio node: You can use it to upload audio that you want to be used in this workflow.
2. Modify the text prompt. You should use keywords to describe the music you want to generate.
3. The strength parameter is used to control the difference from the original audio. The lower the value, the more similar the generated audio will be to the original audio.
4. Click the Run button or use the shortcut Ctrl(cmd) + Enter to execute the audio generation. The audio will be saved to the ComfyUI/output/audio directory.

Audio Inpainting Workflow

Audio inpainting is used to complete or extend existing tracks. You can use it to complete the missing part of music or extend the music to a longer duration.

You need to set where you want to start and end the inpainting.

Download JSON Workflow

Download Input Audio

1. Upload audio to the LoadAudio node.
2. Modify the text prompt. You should use keywords to describe the music you want to generate.
3. (Optional) Modify the duration parameter. It’s 190 by default.
4. (Important) Modify the mask_start and mask_end parameters. You need to set where you want to start and end the inpainting.
5. Click the Run button or use the shortcut Ctrl(cmd) + Enter to execute the audio generation. The audio will be saved to the ComfyUI/output/audio directory.