hunyuan/video-to-video
Transform one video into another style with Tencent Hunyuan Video.
Ace Step is a text-to-music generation model that turns comma-separated style tags and optional lyrics into full songs with vocals, instrumentation, and synchronized lyrics. The model is built for fast iteration, supporting durations from a few seconds up to 4 minutes (240 seconds).
Output format: Audio only / duration 5–240 seconds / stereo / provider-defined sample rate.
| Parameter | Required | Type | Default | Range / Options | Description |
|---|---|---|---|---|---|
| tags* | Yes (*) | string | — | Free text | Comma-separated list of genre, mood, and instrument tags. |
| lyrics | No | string | — | Free text or [inst] / [instrumental] | Vocal content; leave blank for AI-generated lyrics, use [inst] for instrumental. |
| duration | No | integer | 60 | 5 – 240 | Audio length in seconds. |
| seed | No | integer | -1 | -1 – 2147483647 | Random seed for reproducibility; -1 randomizes. |
Ace Step on RunComfy uses time-based billing for generated audio.
| Billing unit | Rate |
|---|---|
| Per second of generated audio | $0.0002 |
Estimated cost examples
| Duration | Approx. cost |
|---|---|
| 30 s | ~$0.006 |
| 60 s (default) | ~$0.012 |
| 120 s | ~$0.024 |
| 240 s (4 min) | ~$0.048 |
1) Open the Ace Step model in RunComfy and reveal the generation panel.
2) Enter style tags such as "lofi, hiphop, chill, mellow piano" to define genre, mood, and instrumentation.
3) Optionally add lyrics; keep verse and chorus sections clearly separated, or use [inst] for an instrumental.
4) Set duration in seconds (5–240); start short to test direction before committing to a full 4-minute render.
5) Lock the seed when you want to compare the impact of tag or lyric changes, or leave it at -1 for variety.
6) Run the generation, preview the result, and download the audio file from your job history.
7) For API use, send the same fields to the Ace Step endpoint on RunComfy; no self-hosting is required.
8) Save promising seeds and tag combinations as presets to keep your sonic direction consistent across a project.
Transform one video into another style with Tencent Hunyuan Video.
Create 1080p cinematic clips from stills with physics-true motion and consistent subjects.
Extend an audio track at the start, end, or both with matching style
Generate cinematic shots guided by reference images with unified control and realistic motion.
Generate premium-quality videos from text prompts with Google Veo 3.
Delivers consistent face animation from a single image using motion-driven synthesis for design and game visualization.
Ace Step is a text-to-music model from acestep-ai that turns style tags and prompts into full audio tracks with melody, rhythm, and vocals. In a text-to-sound workflow on RunComfy, you describe the genre, mood, and structure, and Ace Step generates a coherent musical piece with synchronized lyrics. It is designed for creators who want fast, prompt-driven music generation without manual composition.
Ace Step is best suited for text-to-sound tasks such as generating background music, short song demos, ambient loops, ad jingles, and reference tracks for video or game scenes. It handles style tag control well, so you can steer genre, tempo, and energy with a few descriptors. Vocal and lyric generation also makes it useful for songwriting drafts and creative prototyping.
Compared to many general audio models, Ace Step focuses on fine-grained acoustic fidelity, with attention to dynamic balance, spatial quality, and instrument clarity. The style-tag interface gives technical artists and designers more direct control over genre and energy than free-form-only prompts. Reproducibility through a seed parameter also helps developers iterate consistently on a chosen direction.
Designers, technical artists, video creators, and product teams can use Ace Step text-to-sound generation for trailers, social content, prototype game audio, e-commerce videos, and ad creatives. Developers can wrap it into pipelines that need on-demand soundtracks tied to scene metadata or campaign briefs. Because the model supports both vocals and instrumentals, it covers a wide range of audio needs from a single interface.
Ace Step supports flexible duration, adjustable from a few seconds up to about 4 minutes (240 seconds) per generation. Other constraints such as prompt length, supported audio formats, and tag combinations depend on the current provider configuration, so check the RunComfy parameter panel for exact limits before building around them. Limits may vary by mode or provider settings, and the panel always reflects the live values for the text-to-sound endpoint.
You can prototype Ace Step in the RunComfy AI Playground Web UI by adjusting style tags, prompts, duration, and seed until the text-to-sound output matches your target. Once the configuration is stable, call the same Ace Step model through the RunComfy API with identical parameters to automate generation from your backend or content pipeline. This keeps creative iteration in the browser and production runs in code, without changing the underlying model behavior.
Ace Step generations consume usd / credits from your RunComfy balance, and based on available provider information the model is billed per second at $0.0002. New users typically get a free trial usd amount to experiment, after which usage follows the Generation rules shown on the model page. For the most current rates and any mode-specific differences, refer to the Generation section of the Ace Step page on RunComfy.
RunComfy provides access to the Ace Step model and the workflow to generate audio, but commercial usage rights for the generated music depend on the license from the original model author and provider (acestep-ai). Before releasing tracks in commercial products, ads, films, or games, review the official Ace Step license and any provider terms to confirm allowed use cases. If anything is unclear, you can reach out to hi@runcomfy.com for guidance on platform-side questions.
RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.