chirp-v4 is the Suno V4 music generation model, suitable for turning theme descriptions or existing lyrics into listenable songs, as well as creating instrumental music. It offers both inspiration and custom creation modes, making it suitable for short songs, event music drafts, and lyric demos. On this platform, inputs can be organized according to the specified V4 text limits, and audio and song information can be obtained through task results.
Clarify capacity, inputs and outputs, and invocation methods before choosing a model.
The above figures are this platform's V4 usage specifications. Creation parameters and final output performance should be distinguished according to the corresponding model and operation.
Learn what chirp-v4 can bring to your work.
Inspiration mode is suitable for creative starting points when lyrics have not yet been written. Use brief text to describe the song's theme, genre, and mood, set custom=false, then initiate generation through prompt; choose instrumental when background music is needed. It lets creation begin with textual intent rather than requiring instrument recordings first.
Custom mode is suitable for projects that already have verses, choruses, or campaign copy. Set custom=true, submit lyric, title, and style, and express lyrical content separately from musical direction. V4's style space is relatively compact, so prioritize the core genre, main instruments, and mood, and avoid crowding the input with lengthy descriptions.
Generation results center on an audio URL and can also include a song ID, title, lyrics, status, and actual duration, making it easier for applications to build preview and archiving workflows. For background processing, use async=true to obtain task_id, or configure callback_url to receive results, separating music tasks from front-end interactions.
Start with specific tasks to find where the model can be effective.
Enter the event theme, audience, and desired mood, then use inspiration mode to create a draft for listening; when you have a defined slogan or chorus, switch to custom lyrics. Deliver the audio along with the corresponding text for the planning team to discuss musical direction. Brand names, key slogans, and sung content should be confirmed item by item after listening.
Organize completed lyrics into sections such as verses and choruses, add a brief style description, and generate a vocal version for discussion. This is suitable for checking rhythmic feel, section length, and emotional progression after words are set to music. The finished result can serve as a reference for revising lyrics, rather than treating the returned lyrics field as a word-for-word transcription of the audio.
Describe musical moods such as upbeat, soothing, or tense based on the video visuals, choose instrumental music, and create candidate tracks for rough-cut listening. Save the audio and its actual duration, then match it to visual cues in editing software. This is suitable for exploring music directions, but precise cueing, fixed durations, and final editing still need to be completed in post-production.
Choose based on task complexity, input materials, and expected results.
If existing applications already organize inputs around the lyric and style constraints of chirp-v4, or if tasks mainly involve short songs and music drafts, you can continue choosing V4. Compared with V3.5's 120-second tier, V4's 150-second tier provides more length; this difference is for planning content length, does not mean every generation reaches the limit, and does not constitute a comparison of audio quality or speed.
When lyrics exceed 3000 characters, style descriptions need more space, or male and female vocal preference control is needed, chirp-v4-5 is more suitable. V4.5 has lyric and style limits of 5000 and 1000 characters respectively, with a duration tier of 4 minutes. Do not simply add new-version parameters to V4 as a substitute for switching models; fields in a shared interface do not mean that all versions produce the same results.
From a small-scale task to formal integration.
Define objectives, required inputs, and output requirements, using real business examples as a starting point.
Open the trial page, confirm the parameters supported by this entry point, then submit a small-scale task to review the results.
Keep the complete model ID, use the request format specified in the documentation, and confirm billing rules on the Pricing page.
Before formal use, understand output quality and capability boundaries.
Answers to common questions about using chirp-v4.
Submit model=chirp-v4, action=generate, and custom=false to POST /suno/audios, and describe the theme and musical direction in prompt, with a maximum length of 500 characters. If you want to create instrumental music without vocals, set instrumental=true, then retrieve the completed audio result.
Use custom=true, put the lyrics in lyric, the title in title, and the musical direction in style. V4 limits these three fields to 3000, 80, and 200 characters respectively. It is recommended to distinguish lyric content from style instructions and retain clear sections for easier section-by-section listening and editing after generation.
You cannot treat the desired duration as an exact guarantee. The platform lists a maximum V4 duration of 150 seconds, but actual works may be shorter. Custom generation (custom=true, action=generate) can submit duration as a desired duration hint, but it may be ignored and does not guarantee the target length. Read duration in the result and listen to it; deliverables requiring a fixed number of seconds can be trimmed after generation.
Do not rely on vocal_gender to achieve this control with V4, as this parameter applies starting from V4.5. If male or female voice preference is an important selection criterion, choose chirp-v4-5 or an applicable later model; even among supported models, it is preference guidance rather than a guarantee of strict adherence.
After setting async=true, first save the returned task_id, then query the task to retrieve its status and result; you can also configure callback_url to receive completion notifications. Once the song is complete, read audio_url, id, and duration in data, download the audio, and save the metadata for subsequent listening and archiving.
Model information · Updated: 2026-10-01. See the API and pricing sections for request parameters and billing rules.
Start with a clear goal and assess whether it suits your work based on actual results.
