MiniMax Music 2.6 turns musical direction into a song or instrumental track. Describe genre, mood, instruments, tempo and vocal character in the prompt, then supply your own lyrics, ask the model to develop lyrics from the topic, or choose instrumental music. Section tags and arrangement instructions help express how the piece should develop; sample rate, bitrate and MP3/WAV/PCM format settings prepare the returned audio for your workflow. Listen to the full result to review words, musical structure and the sound before using it in a project.
MiniMax Music 2.6 is developed by MiniMax. Vidgo provides access to the model. Model names and trademarks belong to their respective owners.Open the music task and choose custom lyrics when you have words to sing, automatic lyrics when your topic should become a song, or instrumental mode when the result should be music without vocals. Write the musical brief separately from the lyrics, select audio settings and review the per-generation estimate. These are input modes of the same API task and use the same existing model card.
Use prompt for the musical direction and lyrics for the words and section tags. For automatic writing, keep lyrics empty and enable lyrics_optimizer with vocal mode. For instrumental music, enable is_instrumental, keep lyrics empty and set lyrics_optimizer to false. All modes use the same task; the workflow description below comes directly from the model catalog.
| Model / task | Input and workflow |
|---|---|
| MiniMax Music 2.6 Text to Audio | Create songs and instrumentals from style prompts, with custom or automatic lyrics and configurable audio output. |
The recorded examples include Mandarin indie folk with supplied lyrics, an automatically written English funk-soul song and an instrumental fantasy score. Listen to the separate results and inspect the original lyrics and settings to understand each creation mode. The displayed settings are the actual MP3 configuration used in the records; other supported formats can be selected in the task.
Write genre, mood, tempo, meter, instruments and vocal direction in a 10–2,000-character prompt. The reed-bay record specifies 82 BPM, fingerpicked guitar, warm female alto and a gradually opening chorus. Listen for how those musical instructions shape the generated song.
[Verse] 我把脚步放轻在木桥上 昨夜的薄霜还留在草尖 空了一个冬天的芦苇湾 有一声鸣叫越过了水面 [Pre Chorus] 合上那本写满日期的手册 抬头看见风有了形状 [Chorus] 你们回来了 掠过这片湾 把长长的远方折成一双翅膀 我站在岸边 不急着呼喊 让春天先落在安静的水上 [Bridge] 去年弯下的苇秆又慢慢站起 小小的倒影接住整片天 我在空白一页写下今天 有些等待终于不必计算 [Chorus] 你们回来了 掠过这片湾 把长长的远方折成一双翅膀 我站在岸边 不急着呼喊 让春天先落在安静的水上 [Outro] 木桥轻轻响 风吹过衣裳 芦苇湾留住了第一缕暖阳
Place up to 3,500 characters of words in lyrics and use tags such as [Verse], [Chorus] and [Bridge] to express structure. Keep instrumental and automatic-lyrics flags false for the custom-lyrics path. The displayed reed-bay text is the original supplied lyric input, preserving its Mandarin language and section order.
For automatic lyrics, describe the story, language, vocal character and arrangement in prompt, keep lyrics empty and set lyrics_optimizer true with is_instrumental false. The rooftop-garden record requests an original English funk-soul song, a memorable title chorus and a relaxed male tenor.
Enable instrumental mode, keep lyrics empty and lyrics optimization false, then describe instrumentation, rhythm, mood and musical development. The buried-library record specifies a chamber-orchestra exploration score with cello, bass clarinet and celesta, followed by a brighter discovery section.
{
"is_instrumental": false,
"lyrics_optimizer": false,
"audio_setting": {
"sample_rate": 44100,
"bitrate": 256000,
"format": "mp3"
}
}Set the supported sample rate, bitrate and MP3, WAV or PCM format in audio_setting to match the next step of your workflow. The shown request uses 44,100 Hz, 256,000 bits per second and MP3. Check the returned file and audition the track in the software where you plan to use it.
The reed-bay request supplies complete Mandarin lyrics and asks for intimate indie folk that opens into a spacious chorus. The buried-library request supplies an instrumental exploration brief with changing orchestration and a gentle coda. Read the original musical direction, lyrics when supplied and actual mode/audio settings, then listen for lyrical delivery, instrumentation, section changes and the ending in each corresponding track.
Original musical direction: Mandarin indie folk song, 82 BPM, 4/4, D major. A wetland observer watches migrating birds return to a reed-lined bay after winter. Warm intimate female alto, natural breaths, clear Mandarin diction, understated emotion rather than theatrical belting. Fingerpicked acoustic guitar and a close dry vocal begin alone; cello enters in the pre-chorus; restrained brushed drums and rounded bass support a spacious melodic chorus. Bridge briefly strips back to voice and cello, then a final fuller chorus opens like wings over water. Soft resolved guitar outro, no abrupt cutoff. Keep the supplied Chinese lyrics and section order. Organic recording, clear vocal above the instruments, warm separated low end. English meaning of the supplied Mandarin lyrics: [Verse] I tread softly across the wooden bridge Last night's thin frost still rests on the grass tips The reed-lined bay was empty all winter Now a birdcall crosses the water [Pre Chorus] I close the notebook filled with dates And look up to see the wind take shape [Chorus] You have returned, skimming over this bay Folding the long distance into a pair of wings I stand on the shore, in no hurry to call out Let spring settle first on the quiet water [Bridge] The reeds that bent last year slowly rise again A tiny reflection holds the whole sky On a blank page I write today's date Some waiting finally needs no counting [Chorus] You have returned, skimming over this bay Folding the long distance into a pair of wings I stand on the shore, in no hurry to call out Let spring settle first on the quiet water [Outro] The wooden bridge creaks softly; wind brushes my clothes The reed-lined bay holds the first ray of warmth
[Verse] 我把脚步放轻在木桥上 昨夜的薄霜还留在草尖 空了一个冬天的芦苇湾 有一声鸣叫越过了水面 [Pre Chorus] 合上那本写满日期的手册 抬头看见风有了形状 [Chorus] 你们回来了 掠过这片湾 把长长的远方折成一双翅膀 我站在岸边 不急着呼喊 让春天先落在安静的水上 [Bridge] 去年弯下的苇秆又慢慢站起 小小的倒影接住整片天 我在空白一页写下今天 有些等待终于不必计算 [Chorus] 你们回来了 掠过这片湾 把长长的远方折成一双翅膀 我站在岸边 不急着呼喊 让春天先落在安静的水上 [Outro] 木桥轻轻响 风吹过衣裳 芦苇湾留住了第一缕暖阳
{
"is_instrumental": false,
"lyrics_optimizer": false,
"audio_setting": {
"sample_rate": 44100,
"bitrate": 256000,
"format": "mp3"
}
}Instrumental fantasy exploration score, 96 BPM, 6/8, E minor with a luminous G major discovery section. A lone archivist descends into a sealed underground library and unfolds an ancient star atlas. Begin with a sparse low-cello ostinato, breathy bass clarinet and isolated celesta notes, mysterious and curious. Gradually add pizzicato strings and soft low frame drums with precise attacks and a clean separated bass register. At the discovery, broaden the melody in flute and warm strings, shimmering celesta above, awe rather than battle or horror. Briefly pause the percussion before the main motif returns, then resolve gently to a quiet cello and celesta coda. Acoustic chamber-orchestra character, expressive dynamic contrast, no vocals, no choir, no spoken words, no sound effects, no abrupt cutoff.
{
"is_instrumental": true,
"lyrics_optimizer": false,
"audio_setting": {
"sample_rate": 44100,
"bitrate": 256000,
"format": "mp3"
}
}Prepare original lyrics, a song topic or an instrumental arrangement brief. Describe musical style, emotional progression and the intended listening context, then choose the corresponding mode and audio format. Reviewed tracks can support song demos, narrative score concepts, creator background music, instrumental arrangements, theme-song prototypes and music features in applications.
Prepare words organized into verses, choruses and a bridge, then describe genre and vocal character separately. Generate and listen to the song to review lyrical delivery, melody and arrangement before developing the composition further.
Describe the scene, emotional arc, instruments and transitions in an instrumental brief. Listen for how the music develops through the scene and use the result to explore a game, film or story atmosphere.
Prepare a clear mood, tempo and instrumentation brief suited to the planned video or presentation. Generate an instrumental track and review its density, dynamics and ending alongside the content it will accompany.
Describe how instruments enter, respond and resolve through the piece, including rhythm and section contrast. Audition the result to explore musical textures or an arrangement direction for a larger composition.
Describe an original story, language, title and vocal style, then enable automatic lyrics. Listen to the generated words, chorus and instrumental character to assess a theme-song prototype or creative pitch.
Collect a user’s musical direction and either lyrics or a creation-mode choice. Submit the documented request and retrieve the finished audio by task ID, then let the application play or save the result in its intended workflow.
MiniMax Music 2.6 uses the current per-generation price shown in the dynamic table. The listed task has one generation tier covering its creation modes and supported audio settings. Open the music task to review the complete amount before submitting, and multiply it by the number of planned requests when budgeting for variations.
| Model | Provider | Workflow | Input material | Audio output | Starting rate |
|---|---|---|---|---|---|
| MiniMax Music 2.6 Text to Audio | MiniMax | Text to Audio | Music prompt; optional lyrics | MP3 / WAV / PCM | $0.1 / generation |
Test your creative brief online, then bring the verified request into your application.
Open the music task, enter genre, mood, instruments, tempo and vocal direction, and configure the selected creation mode. Add your lyrics when using custom-lyrics mode and choose sample rate, bitrate and format. Review the estimate, generate and listen through the whole track to check words, arrangement, transitions and the ending.
Choose a task, inspect its settings and cost, and follow the result from an online test to an authenticated API request.
Choose custom lyrics for words you have written, automatic lyrics for a topic-led vocal song, or instrumental mode for music without a sung part. All three choices use the same music task. Describe musical direction in prompt, prepare lyrics when required and choose the audio configuration for the returned file.
Open the music task, enter genre, mood, instruments, tempo and vocal direction, and configure the selected creation mode. Add your lyrics when using custom-lyrics mode and choose sample rate, bitrate and format. Review the estimate, generate and listen through the whole track to check words, arrangement, transitions and the ending.
Use the current per-generation amount and the number of song or instrumental variations you plan to request. Review the full estimate in the music task before submitting and keep the task ID with each result when comparing variations.
Read the actual musical brief, supplied lyrics and mode settings while listening to the result. Use the recorded examples to understand how genre, instruments, vocal direction and song structure are expressed in a request.
Retain the task ID returned by the API to query status and retrieve completed file URLs. Use your online generation history to review earlier results and the creative direction you tested.
Use the documentation linked from each task for its exact model identifier, accepted inputs, authentication and response format. Integrate the same task you tested online so its media and settings match the request.
MiniMax Music 2.6 is a music-generation model on Vidgo for songs and instrumental tracks. One task accepts a musical prompt with custom lyrics, automatic lyric writing or instrumental mode, and returns generated audio using the selected output configuration.
Read music-model introductions, API guides and platform comparisons to understand creative inputs, request integration and cost planning.
