Mureka Song API, Mureka Instrumental API, Mureka Lyrics & Podcast API: Pricing, Documentation

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/mureka-song-prompt/v1/prompt-to-song' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"prompt": "An upbeat summer song about a road trip with friends, with a catchy chorus", "styles": ["pop", "rock"], "model": "mureka-9.5"}'

Callback Payload (success)

{
  "request_id": "mureka-song-prompt_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-song-prompt",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Failure callback shape

{
  "request_id": "mureka-song-prompt_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "mureka-song-prompt",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll while the request is processing (status PROCESSING), plus the final COMPLETED/ERROR callback.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Good to know

• Each request returns one song. batch_size, num_images, number_of_images, n, count, num_instrumentals, num_songs and num_outputs are rejected — send one request per song.
• This API has no length setting: the model sets the length of the song, and the length does not change the amount billed. duration, seconds, num_frames, frames_per_second and music_length_ms are rejected.
• Send every field at the top level of the JSON body; parameters, content and text are rejected.
• Choose the model version with model: mureka-7.6, mureka-o2, mureka-8, mureka-9 or mureka-9.5; mureka-9.5 applies when you leave it out. Each version has its own rate per song.
• To sing your own lyrics, use Lyrics to Song: POST https://gateway.pixazo.ai/mureka-song/v1/lyrics-to-song. lyrics and gender are rejected here; describe the voice in prompt.
• Send 1 to 13 genres in styles, written exactly as listed (see Styles below).
• IDs from Mureka's own upload and voice-clone services are not offered: reference_id, vocal_id and melody_id are rejected.
• Use the JSON types shown: prompt and model are strings and styles is a list of strings, written exactly as listed.
• Omit a field instead of sending null.
• Results are always returned as a URL in output.media_url; sync_mode is accepted only as false.
• Webhooks are set with the X-Webhook-URL / X-Webhook-Mode headers; webhook_url, webhook_mode and callback_url in the JSON body are rejected.
• stream is not supported: the result is delivered when the job completes.

Styles

Send 1 to 13 genres in styles, written exactly as listed: pop, rock, jazz, r&b, edm, ambient, folk, latin, k-pop, j-pop, house, gospel, lo-fi. Leave styles out to describe the genre in prompt only.

Example: a prompt with two genres

{
  "prompt": "A calm late-night song about the city lights",
  "styles": [
    "lo-fi",
    "jazz"
  ],
  "model": "mureka-9.5"
}

Prompting Guide

• Describe the song in prompt, up to 2000 characters: what it is about, the mood, the genre, the instruments and the vocals.
• Add genres in styles to steer the genre (see Styles above).
• To sing your own words, use Lyrics to Song instead: POST https://gateway.pixazo.ai/mureka-song/v1/lyrics-to-song.

Example prompt

An upbeat summer song about a road trip with friends, with a catchy chorus

Model Versions

Set model to mureka-7.6, mureka-o2, mureka-8, mureka-9 or mureka-9.5; mureka-9.5 applies when you leave it out. Each version is billed at its own rate per song; the pricing panel on this page lists them.

Example: choose a model version

{
  "prompt": "A short, cheerful song for a birthday party",
  "model": "mureka-7.6"
}

Request Headers

Response Handling

Parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. Problems that need the model to run are reported through the status endpoint as status: "ERROR", with a message in error saying what went wrong. Failed requests are not billed — the hold is released in full.

Example error responses

A refused request returns error, a message that joins every problem found, and the same problems one by one in violations.

Missing prompt (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'prompt' is required",
  "violations": [
    "'prompt' is required"
  ]
}

An empty styles list (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'styles' must have at least 1 item (got 0)",
  "violations": [
    "'styles' must have at least 1 item (got 0)"
  ]
}

A model version that does not exist (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'model' must be one of: \"mureka-7.6\", \"mureka-o2\", \"mureka-8\", \"mureka-9\", \"mureka-9.5\" (got \"mureka-10\")",
  "violations": [
    "'model' must be one of: \"mureka-7.6\", \"mureka-o2\", \"mureka-8\", \"mureka-9\", \"mureka-9.5\" (got \"mureka-10\")"
  ]
}

A field Prompt to Song does not accept (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'gender' must be one of:  (got \"male\"); 'gender' is not accepted by prompt-to-song. Describe the voice in 'prompt'.",
  "violations": [
    "'gender' must be one of:  (got \"male\")",
    "'gender' is not accepted by prompt-to-song. Describe the voice in 'prompt'."
  ]
}

More than one song per request (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'num_songs' must be one of:  (got 2); 'num_songs' is not supported: each request returns one song. Send one request per song.",
  "violations": [
    "'num_songs' must be one of:  (got 2)",
    "'num_songs' is not supported: each request returns one song. Send one request per song."
  ]
}

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED or ERROR, then download the song from output.media_url. Treat any status other than QUEUED, PROCESSING or COMPLETED as a failure.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY

cURL example

curl 'https://gateway.pixazo.ai/v2/requests/status/mureka-song-prompt_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "mureka-song-prompt_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-song-prompt",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

Typical workflow

1. Send the request to POST https://gateway.pixazo.ai/mureka-song-prompt/v1/prompt-to-song.
2. Save the request_id from the 202 response.
3. Poll GET /v2/requests/status/{request_id} every 10 seconds or so, or use a Webhook instead.
4. When status is COMPLETED, download the song from output.media_url.

How Billing Is Measured

Billing is measured per song: each request is billed once, at the rate of the model version you choose with model (mureka-9.5 when you leave it out). Prompt to Song and Lyrics to Song have separate rates. The length of the song does not change the amount. The amount is placed as a hold when the request is submitted and released in full if the request fails. The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/mureka-song/v1/lyrics-to-song' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"lyrics": "[Verse]\nMorning light on an empty street\nCoffee warm and a steady beat\n[Chorus]\nWe keep on walking, side by side\nNothing to lose and nothing to hide", "prompt": "indie pop, warm, acoustic guitar", "model": "mureka-9.5"}'

Callback Payload (success)

{
  "request_id": "mureka-song_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-song",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Failure callback shape

{
  "request_id": "mureka-song_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "mureka-song",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll while the request is processing (status PROCESSING), plus the final COMPLETED/ERROR callback.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Good to know

• Each request returns one song. batch_size, num_images, number_of_images, n, count, num_instrumentals, num_songs and num_outputs are rejected — send one request per song.
• This API has no length setting: the model sets the length of the song, and the length does not change the amount billed. duration, seconds, num_frames, frames_per_second and music_length_ms are rejected.
• Send every field at the top level of the JSON body; parameters, content and text are rejected.
• Choose the model version with model: mureka-7.6, mureka-o2, mureka-8, mureka-9 or mureka-9.5; mureka-9.5 applies when you leave it out. Each version has its own rate per song.
• To describe a song instead of sending lyrics, use Prompt to Song: POST https://gateway.pixazo.ai/mureka-song-prompt/v1/prompt-to-song. styles is rejected here; describe the style in prompt.
• IDs from Mureka's own upload and voice-clone services are not offered: reference_id, vocal_id and melody_id are rejected. Use gender to ask for a female or male voice.
• Use the JSON types shown: every field is a string, and model and gender take exactly the values listed.
• Omit a field instead of sending null.
• Results are always returned as a URL in output.media_url; sync_mode is accepted only as false.
• Webhooks are set with the X-Webhook-URL / X-Webhook-Mode headers; webhook_url, webhook_mode and callback_url in the JSON body are rejected.
• stream is not supported: the result is delivered when the job completes.
• Do not begin lyrics with http://, https:// or data: — such a value is treated as a file to download, and the request is refused at submit with nothing charged.

Prompting Guide

• Write the words to sing in lyrics, up to 5000 characters. Mark the parts of the song with section tags on their own lines, such as [Verse] and [Chorus], as in the example below.
• Describe the musical style in prompt, up to 1024 characters: genre, mood, instruments and the kind of vocals.
• Use gender to ask for a female or male voice.

Example lyrics with section tags

[Verse]
Morning light on an empty street
Coffee warm and a steady beat
[Chorus]
We keep on walking, side by side
Nothing to lose and nothing to hide

Example: ask for a female voice

{
  "lyrics": "[Verse]\nMorning light on an empty street\nCoffee warm and a steady beat\n[Chorus]\nWe keep on walking, side by side\nNothing to lose and nothing to hide",
  "prompt": "acoustic ballad, gentle",
  "gender": "female",
  "model": "mureka-9.5"
}

Model Versions

Set model to mureka-7.6, mureka-o2, mureka-8, mureka-9 or mureka-9.5; mureka-9.5 applies when you leave it out. Each version is billed at its own rate per song; the pricing panel on this page lists them.

Example: choose a model version

{
  "lyrics": "[Verse]\nMorning light on an empty street\nCoffee warm and a steady beat\n[Chorus]\nWe keep on walking, side by side\nNothing to lose and nothing to hide",
  "model": "mureka-7.6"
}

Request Headers

Response Handling

Parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. Problems that need the model to run are reported through the status endpoint as status: "ERROR", with a message in error saying what went wrong. Failed requests are not billed — the hold is released in full.

Example error responses

A refused request returns error, a message that joins every problem found, and the same problems one by one in violations.

Missing lyrics (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'lyrics' is required",
  "violations": [
    "'lyrics' is required"
  ]
}

A model version that does not exist (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'model' must be one of: \"mureka-7.6\", \"mureka-o2\", \"mureka-8\", \"mureka-9\", \"mureka-9.5\" (got \"mureka-10\")",
  "violations": [
    "'model' must be one of: \"mureka-7.6\", \"mureka-o2\", \"mureka-8\", \"mureka-9\", \"mureka-9.5\" (got \"mureka-10\")"
  ]
}

A length setting (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'duration' must be one of:  (got 120); 'duration' is not accepted: this API has no length setting; the model sets the length of the song.",
  "violations": [
    "'duration' must be one of:  (got 120)",
    "'duration' is not accepted: this API has no length setting; the model sets the length of the song."
  ]
}

More than one song per request (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'n' must be one of:  (got 2); 'n' is not supported: each request returns one song. Send one request per song.",
  "violations": [
    "'n' must be one of:  (got 2)",
    "'n' is not supported: each request returns one song. Send one request per song."
  ]
}

An ID from Mureka's own services (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'vocal_id' must be one of:  (got \"1234567890\"); 'vocal_id' is not supported on this API: it needs an ID from Mureka's own voice clone service, which this API does not offer. Use 'gender' to ask for a female or male voice.",
  "violations": [
    "'vocal_id' must be one of:  (got \"1234567890\")",
    "'vocal_id' is not supported on this API: it needs an ID from Mureka's own voice clone service, which this API does not offer. Use 'gender' to ask for a female or male voice."
  ]
}

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED or ERROR, then download the song from output.media_url. Treat any status other than QUEUED, PROCESSING or COMPLETED as a failure.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY

cURL example

curl 'https://gateway.pixazo.ai/v2/requests/status/mureka-song_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "mureka-song_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-song",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

Typical workflow

1. Send the request to POST https://gateway.pixazo.ai/mureka-song/v1/lyrics-to-song.
2. Save the request_id from the 202 response.
3. Poll GET /v2/requests/status/{request_id} every 10 seconds or so, or use a Webhook instead.
4. When status is COMPLETED, download the song from output.media_url.

How Billing Is Measured

Billing is measured per song: each request is billed once, at the rate of the model version you choose with model (mureka-9.5 when you leave it out). Lyrics to Song and Prompt to Song have separate rates. The length of the song does not change the amount. The amount is placed as a hold when the request is submitted and released in full if the request fails. The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/mureka-instrumental/v1/text-to-music' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"prompt": "Warm lo-fi piano with soft drums and a slow, relaxed groove", "model": "mureka-9.5"}'

Callback Payload (success)

{
  "request_id": "mureka-instrumental_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-instrumental",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Failure callback shape

{
  "request_id": "mureka-instrumental_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "mureka-instrumental",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll while the request is processing (status PROCESSING), plus the final COMPLETED/ERROR callback.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Good to know

• Each request returns one track. batch_size, num_images, number_of_images, n, count, num_instrumentals, num_songs and num_outputs are rejected — send one request per track.
• This API has no length setting: the model sets the length of the track, and the length does not change the amount billed. duration, seconds, num_frames, frames_per_second and music_length_ms are rejected.
• Send every field at the top level of the JSON body; parameters, content and text are rejected.
• Choose the model version with model: mureka-7.6, mureka-8, mureka-9 or mureka-9.5; mureka-9.5 applies when you leave it out. Each version has its own rate per track.
• This API makes instrumental music: lyrics, styles, gender and vocal_id are rejected. To add vocals and lyrics, use Lyrics to Song: POST https://gateway.pixazo.ai/mureka-song/v1/lyrics-to-song.
• IDs from Mureka's own upload services are not offered: reference_id, melody_id and instrumental_id are rejected; describe the music in prompt.
• Use the JSON types shown: prompt and model are strings, and model takes exactly the values listed.
• Omit a field instead of sending null.
• Results are always returned as a URL in output.media_url; sync_mode is accepted only as false.
• Webhooks are set with the X-Webhook-URL / X-Webhook-Mode headers; webhook_url, webhook_mode and callback_url in the JSON body are rejected.
• stream is not supported: the result is delivered when the job completes.

Prompting Guide

• Describe the music in prompt, up to 1024 characters: genre, mood, tempo and instruments.
• This API makes instrumental music. To add vocals and lyrics, use Lyrics to Song: POST https://gateway.pixazo.ai/mureka-song/v1/lyrics-to-song.

Example prompt

Warm lo-fi piano with soft drums and a slow, relaxed groove

Model Versions

Set model to mureka-7.6, mureka-8, mureka-9 or mureka-9.5; mureka-9.5 applies when you leave it out. Each version is billed at its own rate per track; the pricing panel on this page lists them.

Example: choose a model version

{
  "prompt": "Bright acoustic guitar with light percussion",
  "model": "mureka-7.6"
}

Request Headers

Response Handling

Parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. Problems that need the model to run are reported through the status endpoint as status: "ERROR", with a message in error saying what went wrong. Failed requests are not billed — the hold is released in full.

Example error responses

A refused request returns error, a message that joins every problem found, and the same problems one by one in violations.

Missing prompt (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'prompt' is required",
  "violations": [
    "'prompt' is required"
  ]
}

A model version this API does not offer (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'model' must be one of: \"mureka-7.6\", \"mureka-8\", \"mureka-9\", \"mureka-9.5\" (got \"mureka-o2\")",
  "violations": [
    "'model' must be one of: \"mureka-7.6\", \"mureka-8\", \"mureka-9\", \"mureka-9.5\" (got \"mureka-o2\")"
  ]
}

An ID from Mureka's own services (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'instrumental_id' must be one of:  (got \"1234567890\"); 'instrumental_id' is not supported on this API: it needs an ID from Mureka's own file upload service, which this API does not offer. Describe the music in 'prompt'.",
  "violations": [
    "'instrumental_id' must be one of:  (got \"1234567890\")",
    "'instrumental_id' is not supported on this API: it needs an ID from Mureka's own file upload service, which this API does not offer. Describe the music in 'prompt'."
  ]
}

A length setting (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'duration' must be one of:  (got 60); 'duration' is not accepted: this API has no length setting; the model sets the length of the track.",
  "violations": [
    "'duration' must be one of:  (got 60)",
    "'duration' is not accepted: this API has no length setting; the model sets the length of the track."
  ]
}

More than one track per request (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'num_instrumentals' must be one of:  (got 2); 'num_instrumentals' is not supported: each request returns one instrumental track. Send one request per track.",
  "violations": [
    "'num_instrumentals' must be one of:  (got 2)",
    "'num_instrumentals' is not supported: each request returns one instrumental track. Send one request per track."
  ]
}

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED or ERROR, then download the track from output.media_url. Treat any status other than QUEUED, PROCESSING or COMPLETED as a failure.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY

cURL example

curl 'https://gateway.pixazo.ai/v2/requests/status/mureka-instrumental_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "mureka-instrumental_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-instrumental",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

Typical workflow

1. Send the request to POST https://gateway.pixazo.ai/mureka-instrumental/v1/text-to-music.
2. Save the request_id from the 202 response.
3. Poll GET /v2/requests/status/{request_id} every 10 seconds or so, or use a Webhook instead.
4. When status is COMPLETED, download the track from output.media_url.

How Billing Is Measured

Billing is measured per track: each request is billed once, at the rate of the model version you choose with model (mureka-9.5 when you leave it out). The length of the track does not change the amount. The amount is placed as a hold when the request is submitted and released in full if the request fails. The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/mureka-podcast/v1/text-to-podcast' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"conversations": [{"text": "Welcome back to the show. Today we are talking about simple ways to sleep better.", "voice": "Luna"}, {"text": "Thanks, Luna. The easiest one first: go to bed at the same time every night.", "voice": "Jake"}]}'

Callback Payload (success)

{
  "request_id": "mureka-podcast_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-podcast",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Failure callback shape

{
  "request_id": "mureka-podcast_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "mureka-podcast",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll while the request is processing (status PROCESSING), plus the final COMPLETED/ERROR callback.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Good to know

• Each request returns one podcast. batch_size, num_images, number_of_images, n and count are rejected — send one request per podcast.
• The length of the podcast follows the script in conversations. duration, seconds, num_frames, frames_per_second and music_length_ms are rejected.
• Send every field at the top level of the JSON body; parameters, content and text are rejected.
• Send the script in conversations: 1 to 10 turns, each with text (up to 400 characters) and voice (Ethan, Victoria, Jake, Luna or Emma).
• voice, model and prompt are rejected at the top level: set a voice on each turn, and send the script as conversations.
• The length of each turn and the voice names are checked when the podcast is generated: a turn over 400 characters or an unknown voice fails the request with status: "ERROR", and nothing is charged.
• Use the JSON types shown: conversations is a list of objects whose text and voice are strings.
• Omit a field instead of sending null.
• Results are always returned as a URL in output.media_url; sync_mode is accepted only as false.
• Webhooks are set with the X-Webhook-URL / X-Webhook-Mode headers; webhook_url, webhook_mode and callback_url in the JSON body are rejected.
• stream is not supported: the result is delivered when the job completes.
• Do not begin the text of a turn with http://, https:// or data: — such a value is treated as a file to download, and the request is refused at submit with nothing charged.

Script and Voices

• Send the script in conversations: 1 to 10 turns, in the order they are spoken. Each turn has text (1 to 400 characters) and voice.
• Voices: Ethan, Victoria, Jake, Luna, Emma, written exactly as listed. Give each speaker their own voice.
• A turn over 400 characters or an unknown voice is refused when the podcast is generated (status: "ERROR", nothing charged).

Example: a three-turn script with two voices

{
  "conversations": [
    {
      "text": "Hello and welcome. This week we look at three easy habits for a calmer morning.",
      "voice": "Ethan"
    },
    {
      "text": "I tried the first one all week: no phone for the first ten minutes after waking up.",
      "voice": "Victoria"
    },
    {
      "text": "That is a great place to start. Let us hear how it went.",
      "voice": "Ethan"
    }
  ]
}

Request Headers

Response Handling

Parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. Problems that need the model to run are reported through the status endpoint as status: "ERROR", with a message in error saying what went wrong. Failed requests are not billed — the hold is released in full.

Example error responses

A refused request returns error, a message that joins every problem found, and the same problems one by one in violations.

An empty script (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'conversations' is required",
  "violations": [
    "'conversations' is required"
  ]
}

More than 10 turns (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'conversations' must have at most 10 items (got 11)",
  "violations": [
    "'conversations' must have at most 10 items (got 11)"
  ]
}

A voice at the top level (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'voice' must be one of:  (got \"Luna\"); 'voice' is not accepted at the top level. Set a voice on each turn as conversations[].voice.",
  "violations": [
    "'voice' must be one of:  (got \"Luna\")",
    "'voice' is not accepted at the top level. Set a voice on each turn as conversations[].voice."
  ]
}

A length setting (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'duration' must be one of:  (got 60); 'duration' is not accepted: the length of the podcast follows the script in 'conversations'.",
  "violations": [
    "'duration' must be one of:  (got 60)",
    "'duration' is not accepted: the length of the podcast follows the script in 'conversations'."
  ]
}

A prompt instead of a script (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'conversations' is required; 'prompt' must be one of:  (got \"A podcast about sleep\"); 'prompt' is not accepted. Send the script as 'conversations': a list of turns, each with 'text' and 'voice'.",
  "violations": [
    "'conversations' is required",
    "'prompt' must be one of:  (got \"A podcast about sleep\")",
    "'prompt' is not accepted. Send the script as 'conversations': a list of turns, each with 'text' and 'voice'."
  ]
}

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED or ERROR, then download the podcast from output.media_url. Treat any status other than QUEUED, PROCESSING or COMPLETED as a failure.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY

cURL example

curl 'https://gateway.pixazo.ai/v2/requests/status/mureka-podcast_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "mureka-podcast_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-podcast",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

Typical workflow

1. Send the request to POST https://gateway.pixazo.ai/mureka-podcast/v1/text-to-podcast.
2. Save the request_id from the 202 response.
3. Poll GET /v2/requests/status/{request_id} every 10 seconds or so, or use a Webhook instead.
4. When status is COMPLETED, download the podcast from output.media_url.

How Billing Is Measured

Billing is measured per minute of generated audio. A podcast is billed in whole minutes: its length is rounded up to the next full minute. A hold is placed on your balance when you submit; when the podcast is ready, the charge is set from the length of the generated audio and any unused part of the hold is released. A failed request is not charged. The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/mureka-lyrics/v1/text-to-lyrics' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"prompt": "A song about leaving home for the first time"}'

Callback Payload (success)

{
  "request_id": "mureka-lyrics_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-lyrics",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.json"
    ],
    "media_type": "application/json",
    "title": "The generated title",
    "lyrics": "The generated lyrics"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Failure callback shape

{
  "request_id": "mureka-lyrics_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "mureka-lyrics",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll while the request is processing (status PROCESSING), plus the final COMPLETED/ERROR callback.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Good to know

• Each request returns one set of lyrics. batch_size, num_images, number_of_images, n and count are rejected — send one request per set of lyrics.
• This API has no length setting. duration, seconds, num_frames, frames_per_second and music_length_ms are rejected.
• Send every field at the top level of the JSON body; parameters, content and text are rejected.
• Lyrics generation has no model setting: model is rejected.
• Send the topic in prompt; lyrics is rejected.
• Each result holds one title and one set of lyrics. To turn the lyrics into a song, send them as lyrics to Lyrics to Song: POST https://gateway.pixazo.ai/mureka-song/v1/lyrics-to-song.
• Use the JSON types shown: prompt is a string.
• Omit a field instead of sending null.
• Results are always returned as a URL in output.media_url; sync_mode is accepted only as false.
• Webhooks are set with the X-Webhook-URL / X-Webhook-Mode headers; webhook_url, webhook_mode and callback_url in the JSON body are rejected.
• stream is not supported: the result is delivered when the job completes.

Using the Lyrics

A completed request holds a title in output.title and the lyrics in output.lyrics; the same two fields are also stored as a JSON file at output.media_url[0]. To turn the lyrics into a song, send output.lyrics as lyrics to Lyrics to Song: POST https://gateway.pixazo.ai/mureka-song/v1/lyrics-to-song.

Request Headers

Response Handling

Parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. Problems that need the model to run are reported through the status endpoint as status: "ERROR", with a message in error saying what went wrong. Failed requests are not billed — the hold is released in full.

Example error responses

A refused request returns error, a message that joins every problem found, and the same problems one by one in violations.

Missing prompt (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'prompt' is required",
  "violations": [
    "'prompt' is required"
  ]
}

A prompt longer than 2000 characters (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'prompt' must be at most 2000 characters (got 2001)",
  "violations": [
    "'prompt' must be at most 2000 characters (got 2001)"
  ]
}

A field this API does not accept (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'model' must be one of:  (got \"mureka-9.5\"); 'model' is not accepted: lyrics generation has no model setting.",
  "violations": [
    "'model' must be one of:  (got \"mureka-9.5\")",
    "'model' is not accepted: lyrics generation has no model setting."
  ]
}

More than one set of lyrics per request (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'n' must be one of:  (got 2); 'n' is not supported: each request returns one set of lyrics. Send one request per set.",
  "violations": [
    "'n' must be one of:  (got 2)",
    "'n' is not supported: each request returns one set of lyrics. Send one request per set."
  ]
}

Lyrics sent instead of a topic (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'lyrics' must be one of:  (got \"[Verse]\"); 'lyrics' is not accepted by this API. Send the topic of the song in 'prompt'.",
  "violations": [
    "'lyrics' must be one of:  (got \"[Verse]\")",
    "'lyrics' is not accepted by this API. Send the topic of the song in 'prompt'."
  ]
}

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED or ERROR, then read output.title and output.lyrics (the same two fields are stored as a JSON file at output.media_url[0]). Treat any status other than QUEUED, PROCESSING or COMPLETED as a failure.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY

cURL example

curl 'https://gateway.pixazo.ai/v2/requests/status/mureka-lyrics_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "mureka-lyrics_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-lyrics",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.json"
    ],
    "media_type": "application/json",
    "title": "The generated title",
    "lyrics": "The generated lyrics"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

Typical workflow

1. Send the request to POST https://gateway.pixazo.ai/mureka-lyrics/v1/text-to-lyrics.
2. Save the request_id from the 202 response.
3. Poll GET /v2/requests/status/{request_id} every 10 seconds or so, or use a Webhook instead.
4. When status is COMPLETED, read output.title and output.lyrics (the same two fields are stored as a JSON file at output.media_url[0]).

How Billing Is Measured

Billing is measured per request: each request is billed once, for one title and one set of lyrics. The amount is placed as a hold when the request is submitted and released in full if the request fails. The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/mureka-lyrics-video/v1/song-to-video' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"song_request_id": "mureka-song_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx", "selection_start": 0, "selection_end": 30000, "aspect_ratio": "9:16", "layout": "layout_1"}'

Callback Payload (success)

{
  "request_id": "mureka-lyrics-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-lyrics-video",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp4"
    ],
    "media_type": "video/mp4"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Failure callback shape

{
  "request_id": "mureka-lyrics-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "mureka-lyrics-video",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll while the request is processing (status PROCESSING), plus the final COMPLETED/ERROR callback.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Good to know

• Each request returns one video. batch_size, num_images, number_of_images, n and count are rejected — send one request per video.
• This API has no length setting: the video covers the part of the song you choose with a time range or lyric rows (see below). duration, seconds, num_frames, frames_per_second and music_length_ms are rejected.
• Send every field at the top level of the JSON body; parameters, content and text are rejected.
• Name the song by song_request_id: the request_id you received when you submitted a Lyrics to Song, Prompt to Song or Text to Music (Instrumental) request with the same API key. The song must have completed in the last 28 days. A request made with another API key is never accepted, and songs generated before lyrics videos were available cannot be used.
• song_id is rejected: send song_request_id instead.
• Lyric rows (lyrics_start_row / lyrics_end_row) need a song with lyrics: they are refused for an instrumental track, where selection_start / selection_end select the part to use.
• cover_url needs layout set to one of layout_2 to layout_7; layout_1, the default, has no cover.
• Choose the part of the song: send selection_start and selection_end (milliseconds), or lyrics_start_row and lyrics_end_row. One of the two pairs is required: a request with neither is refused. For the whole song, send selection_start 0 and selection_end at the song's length, rounded down to the whole second. Each pair goes together, and the two kinds cannot be combined.
• Send a cover image as cover_url; cover is rejected.
• Files from Mureka's own upload service are not offered: upload_audio_id and background_id are rejected.
• Use the JSON types shown: lyrics_start_row, lyrics_end_row, selection_start and selection_end are whole numbers (0, not "0"); the other fields are strings.
• Omit a field instead of sending null.
• Results are always returned as a URL in output.media_url; sync_mode is accepted only as false.
• Webhooks are set with the X-Webhook-URL / X-Webhook-Mode headers; webhook_url, webhook_mode and callback_url in the JSON body are rejected.
• stream is not supported: the result is delivered when the job completes.
• Do not begin title with http://, https:// or data: — such a value is treated as a file to download, and the request is refused at submit with nothing charged.
• Options this API does not offer are rejected: show_author_name.

Song, Layout and Selection

• song_request_id is the request_id you received when you submitted the song: a Lyrics to Song, Prompt to Song or Text to Music (Instrumental) request made with the same API key, completed in the last 28 days. Lyric rows need a song with lyrics.
• Choose the template with layout: layout_1, layout_2, layout_3, layout_4, layout_5, layout_6 or layout_7; layout_1 applies when you leave it out and has no cover. A cover_url needs one of layout_2 to layout_7.
• Choose the frame shape with aspect_ratio: 16:9, 9:16, 3:4 or 4:3; 9:16 applies when you leave it out.
• Choose the part of the song for the video: send either selection_start and selection_end in milliseconds (for example a 30-second part), or lyrics_start_row and lyrics_end_row, not both. One of the two is required. For the whole song, send selection_start 0 and selection_end at the song's length in milliseconds, rounded down to the whole second (for example 160000 for a song of 160.2 seconds).

Example: a title, a cover image and lyric rows 1 to 8

{
  "song_request_id": "mureka-song_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "title": "Morning Light",
  "layout": "layout_2",
  "cover_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/doc-assets/images/input.jpg",
  "lyrics_start_row": 1,
  "lyrics_end_row": 8
}

Example: 30 seconds from the middle of the song, in landscape

{
  "song_request_id": "mureka-song_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "selection_start": 60000,
  "selection_end": 90000,
  "aspect_ratio": "16:9"
}

Request Headers

Response Handling

Parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. A cover_url that cannot be downloaded is also refused at submit with a 400, and its hold is released at once: nothing is charged. Problems that need the model to run are reported through the status endpoint as status: "ERROR", with a message in error saying what went wrong. Failed requests are not billed — the hold is released in full.

Example error responses

A refused request returns error, a message that joins every problem found, and the same problems one by one in violations.

Missing song_request_id (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'song_request_id' is required",
  "violations": [
    "'song_request_id' is required"
  ]
}

A song_id sent instead of song_request_id (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'song_request_id' is required; 'song_id' must be one of:  (got \"1234567890\"); 'song_id' is not accepted on this API. Send 'song_request_id': the request_id of a song or instrumental job made with this API key.",
  "violations": [
    "'song_request_id' is required",
    "'song_id' must be one of:  (got \"1234567890\")",
    "'song_id' is not accepted on this API. Send 'song_request_id': the request_id of a song or instrumental job made with this API key."
  ]
}

A value that is not allowed (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'aspect_ratio' must be one of: \"16:9\", \"9:16\", \"3:4\", \"4:3\" (got \"21:9\")",
  "violations": [
    "'aspect_ratio' must be one of: \"16:9\", \"9:16\", \"3:4\", \"4:3\" (got \"21:9\")"
  ]
}

A cover with layout_1 (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'cover_url' needs 'layout' set to one of layout_2 to layout_7: layout_1 (the default) has no cover.",
  "violations": [
    "'cover_url' needs 'layout' set to one of layout_2 to layout_7: layout_1 (the default) has no cover."
  ]
}

No part of the song chosen (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'selection_start' is required unless 'lyrics_start_row' is provided",
  "violations": [
    "'selection_start' is required unless 'lyrics_start_row' is provided"
  ]
}

Half of a selection (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'selection_start' must be sent with 'selection_end' (milliseconds) and cannot be combined with lyric-row selection.",
  "violations": [
    "'selection_start' must be sent with 'selection_end' (milliseconds) and cannot be combined with lyric-row selection."
  ]
}

A length setting (synchronous 400)

{
  "error": "Invalid request parameters",
  "message": "'duration' must be one of:  (got 30); 'duration' is not accepted: to use part of the song, send 'selection_start' and 'selection_end' (milliseconds) or 'lyrics_start_row' and 'lyrics_end_row'.",
  "violations": [
    "'duration' must be one of:  (got 30)",
    "'duration' is not accepted: to use part of the song, send 'selection_start' and 'selection_end' (milliseconds) or 'lyrics_start_row' and 'lyrics_end_row'."
  ]
}

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED or ERROR, then download the video from output.media_url. Treat any status other than QUEUED, PROCESSING or COMPLETED as a failure.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY

cURL example

curl 'https://gateway.pixazo.ai/v2/requests/status/mureka-lyrics-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "mureka-lyrics-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "mureka-lyrics-video",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp4"
    ],
    "media_type": "video/mp4"
  },
  "created_at": "2026-10-09T10:04:54.387Z",
  "completed_at": "2026-10-09 10:06:21"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

Typical workflow

1. Send the request to POST https://gateway.pixazo.ai/mureka-lyrics-video/v1/song-to-video.
2. Save the request_id from the 202 response.
3. Poll GET /v2/requests/status/{request_id} every 10 seconds or so, or use a Webhook instead.
4. When status is COMPLETED, download the video from output.media_url.

How Billing Is Measured

Billing is measured per video: each request is billed once. The layout, aspect ratio, title, cover and the selected part of the song do not change the amount. The amount is placed as a hold when the request is submitted and released in full if the request fails. A song_request_id that is refused is answered before any hold, so nothing is charged. The current rate is shown in the pricing panel on this page.