Suno V6 is the flagship model in the official new-generation music model family, emphasizing understanding of musical language, precise expression, and the coherence of complete works. It is suited for creations with clearly defined musical style, instrumentation, lyrics, and section structure, selected on the platform through chirp-v6. Work is organized around generation, modification, and task results; specific operations follow this platform's API usage.
Submit requests to the public API at api.acedata.cloud using the documented parameters, then use the results in your application.
Input parameters and result formats vary by service. Use the public API for this model and follow its guide for generation, task retrieval and editing operations.
Clarify capacity, inputs and outputs, and calling methods before selecting a model.
The values above are platform input parameter ranges and do not represent native capacity limits; final duration and media content are subject to actual generation results.
Learn what chirp-v6 can bring to your work.
Officially positioned as a more reliable and precise flagship music creation model, V6 is suitable for starting from an established musical direction, describing vocals, instruments, structure, and mood, then checking whether all elements work together consistently throughout the song.
Custom creation can use style_influence to adjust stylistic tendencies, weirdness to explore more experimental expression, and negative_tags to describe styles to avoid. variety controls result diversity, making it suitable for finding different directions around the same lyrics; these controls guide creation rather than locking in musical results item by item.
chirp-v6 is called through a unified audio endpoint, with results organized as a song list containing information such as audio URLs, song identifiers, status, and actual duration. Applications can use asynchronous tasks or callbacks to handle submission, waiting, previewing, and saving separately, making it suitable for both creative tools and content systems that need to generate music in the background.
Start with specific tasks to find where the model can be effective.
Enter a holiday, brand event, or story theme, add mood and musical style, and first generate a song draft that can be previewed. Once the copy is finalized, create the next version with custom lyrics, placing key information in the verses or chorus. The deliverables are audio and corresponding lyric information, making it easier for teams to discuss the direction of expression rather than reviewing text alone.
Describe the music needed around the atmosphere of the visuals, such as an upbeat trip, quiet narration, or tense transition, and choose instrumental creation. You can change the style description and diversity setting, then select a version suitable for editing after previewing different directions. When precise matching to the video length is needed, editing should be based on the actual audio rather than relying only on the expected duration.
Organize original lyrics into sections such as verses and choruses, then submit them for custom generation with a title, genre, and vocal preferences. First check the lyric performance and overall style, then iterate by adjusting experimentalism or style influence. Keeping the song identifier, settings, and preview records for each version helps compare different interpretations and ultimately select the version that better matches the intent of the lyrics.
Choose based on task complexity, input materials, and expected results.
chirp-v6 is the recommended choice for general creation in the v6 family; chirp-v6-wild and chirp-v6-mini are other models called separately, not switches added to chirp-v6. To compare family members, it is recommended to audition them using the same lyrics, style, and evaluation criteria. Officially, the flagship V6 is positioned for precise, coherent work generation, Mini emphasizes speed and efficiency, and Wild emphasizes rich, less predictable creative exploration; actual time and output performance on this platform still need to be compared through auditioning.
When moving from chirp-v5 or chirp-v5-5 to chirp-v6, it is suitable to first create comparison works using the lyrics and style of existing projects. Older models remain compatible, but the maximum duration and text capacity of older versions cannot be directly applied to v6. If an existing work only needs continuation or partial modification, first clarify the editing goal; there is no need to regenerate all content just to switch models.
Arrange the inputs first, then connect them to the corresponding application workflow.
Prepare the target genre, primary instrumentation, section structure, and original lyrics, and specify which elements must be preserved.
Explicitly select chirp-v6. For theme creation, use prompt; for custom creation, organize lyric, style, and title according to this model's limits. Save task_id, retrieve audio_url after waiting for the result to complete, and archive it together with the work ID and creation settings.
Design tasks directly from the following inputs and acceptance priorities.
Use custom mode to create a “warm, forward-moving brand theme song,” clearly specifying vocals, rhythm, verse-chorus changes, and a natural ending, and explicitly select chirp-v6.
Check whether the creative direction is followed, whether the sections are cohesive, and whether the audio is complete; use the actual result as the basis for the next round of revisions, and avoid changing all parameters at once.
Before formal use, understand the output quality and capability scope.
Answers to common questions when using chirp-v6.
Submit a generation request to /suno/audios, specify model=chirp-v6, use custom=false, and write the theme, mood, and genre in prompt. Prompts are limited to 500 characters, so focus on the most important direction; for instrumental music, you can use the instrumental option.
Use custom=true, put the lyrics in lyric, and use title and style to supplement the work's name and musical direction. Verses and choruses can be organized into sections to express structure more easily. If you only have a lyric concept without the full text, you can omit lyric and instead use lyric_prompt to guide lyric generation.
duration cannot be treated as a strict fixed-length instruction. This endpoint accepts integer values from 10–360 seconds for the expected duration of custom generation, but the finished work may deviate from or be shorter than the set value. Before delivery, read the actual duration in the response and listen to it; this parameter range is also not the native maximum song length.
In the generate operation, both custom=true and max_mode=true must be set to enable enhanced generation mode. It is used to improve consistency, not as a master switch that automatically enables all editing capabilities. To adjust diversity, you can additionally use variety and determine through listening whether it meets the creative goal.
You can set async=true to first obtain task_id and then query the task status; you can also configure callback_url to receive completed results. After completion, read audio_url, the song identifier, and the actual duration from the song list, then save the audio. Cover or video fields may be empty and should not be treated as necessary conditions for successful generation.
Choose a development approach or application suited to this model and follow its setup guide.
Python SDK Integration Guide
SDK installation, authentication and examplesTypeScript SDK Integration Guide
SDK installation, authentication and examples
