Compose AI Music from Text with Stable Audio 2.5

Seedance 2.5 IS HERE · Nano Banana 2 & GPT Image 2.5 UNLIMITED UNTIL September 25Upgrade

Overview

Stable Audio 2.5 is a text-to-audio model that generates original music tracks and sound effects from written prompts. If you need background music for a video, a soundscape for a podcast, or an audio bed for an ad, you describe what you want and the model produces it. On Picasso IA you can run it directly in your browser without any setup. The model handles everything from upbeat electronic loops to slow cinematic strings, all from a one-line description.

How It Works

Type a text prompt describing the audio you want: genre, mood, instruments, tempo, and anything else that matters to you.
Set the duration in seconds, up to 190 seconds for longer tracks or shorter for quick clips.
Adjust the number of diffusion steps if you want higher audio quality and are willing to wait a bit longer for the result.
Optionally raise the guidance scale to make the output follow your description more strictly.
Hit generate and download your audio file once it is ready.

Frequently Asked Questions

Do I need programming skills or technical knowledge to use this? No, just open Stable Audio 2.5 on Picasso IA, adjust the settings you want, and hit generate.

Is it free to try? Yes, you can generate audio with Stable Audio 2.5 on Picasso IA without paying. Credits apply for longer sessions but casual use costs nothing to start.

How long does it take to get results? At the default setting of 8 diffusion steps, most tracks are ready in under 30 seconds. Increasing the step count raises quality but also extends generation time proportionally.

What output formats are supported? The model returns a standard audio file you can download and use in any video editor, DAW, or content platform right away.

Can I customize the output quality or style? Yes. The guidance scale controls how closely the audio follows your prompt, and the step count controls detail. Adjusting both gives you real control over the result without needing to rewrite the prompt.

How many times can I run the model? You can generate as many outputs as your credits allow. If a result misses the mark, try a more specific prompt or change the seed to get a fresh variation.

Where can I use the outputs? The generated audio belongs to you once it is downloaded. Use it in videos, games, presentations, or any project that needs original sound.

Credit Cost

Each generation consumes 4 credits

4 credits

or 20 credits for 5 generations

Features

Everything this model can do for you

Full-length audio

Generate tracks up to 190 seconds long, enough for a full intro, verse, or ambient loop.

Text-driven creation

Describe any genre, mood, or instrumentation and the model produces a matching audio file.

Quality control

Increase diffusion steps to trade generation speed for sharper, more detailed audio output.

Prompt adherence

Raise the guidance scale to make outputs follow your description more closely.

Reproducible results

Reuse any seed value to regenerate the exact same audio across multiple sessions.

No downloads needed

Run the model in a browser on Picasso IA without installing software or configuring anything.

Varied output types

Generate full music tracks, ambient sound layers, or short one-shot sound effects from a single prompt.

Ideal for creative and commercial projects

Use Cases

Generate a full 30-second background music loop for a YouTube video by describing the genre and mood in a text prompt

Create custom ambient soundscapes for podcast intros by specifying instruments, tempo, and atmosphere in plain language

Produce sound effects for a game prototype by describing the audio event, such as a tense drone or a victory fanfare

Draft a music bed for a product ad by entering a one-line brief describing energy level, style, and instrumentation

Create short audio clips for social media reels by describing the vibe and letting the model generate a loopable track

Generate placeholder music for a film scene before hiring a composer, using a text description of the emotional tone

Experiment with unfamiliar music genres by describing their characteristics and listening to generated examples instantly

Experimenting with genre-blending audio

Examples

90s

7.0s

Steps: 8

Cfg Scale: 1

Soulful Boom Bap Hip Hop instrumental, Solemn effected Piano, SP-1200, low-key swing drums, sine wave bass, Characterful, Peaceful, Interesting, well-arranged composition, 90 BPM

90s

6.2s

Steps: 8

Cfg Scale: 1

Ambient house, 808 drum machine, 808 kick, claps, shaker, synthesizer, synth bass, modern, futuristic, Dancy, Euphoric, 125 BPM

90s

6.1s

Steps: 8

Cfg Scale: 1

Cinematic Synthwave

90s

6.4s

Steps: 8

Cfg Scale: 1

Lo-fi Funk

90s

5.7s

Steps: 8

Cfg Scale: 1

Post-Rock, Guitars, Drum Kit, Bass, Strings, Euphoric, Up-Lifting, Moody, Flowing, Raw, Epic, Sentimental, 125 BPM

90s

6.4s

Steps: 8

Cfg Scale: 1

Piano melody begins a melancholic journey, full orchestral climax, the swells of the orchestral instrumentals

90s

6.5s

Steps: 8

Cfg Scale: 1

Soulful Boom Bap Hip Hop instrumental, Solemn effected Piano, SP-1200, low-key swing drums, sine wave bass, Characterful, Peaceful, Interesting, well-arranged composition, 90 BPM

90s

5.8s

Steps: 8

Cfg Scale: 1

Pop, Pop-Electronic, Ballad, Billboard, Drum Machine, Bass, Lush Synthesizer Pads, Synthesizer Arp, Synth Bass, Vocal Sample Chops, Percussion, Honest, Heart-Felt, Melancholic, Vibe, Cool, Modern, Atmospheric, well-arranged composition, 115 BPM