Skip to main content
POST
/
proxy
/
sonilo
/
t2m
/
generate
Generate music from text prompt
curl --request POST \
  --url https://api.comfy.org/proxy/sonilo/t2m/generate \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form duration=180 \
  --form 'prompt=<string>'
{
  "copy_index": 1,
  "display_tags": [
    "<string>"
  ],
  "prompt_index": 1,
  "stream_index": 1,
  "title": "<string>",
  "type": "title",
  "summary": "<string>"
}

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

multipart/form-data
​
duration
integer
required

Desired duration of the output track in seconds.

Required range: 1 <= x <= 360
​
prompt
string
required

Text prompt describing the desired music. Max length 1000 characters.

Required string length: 1 - 1000

Response

OK - Streaming NDJSON response with audio generation events

A single NDJSON event from the Sonilo streaming response. Additional event types beyond those listed may appear; unknown types should be ignored by clients.

​
copy_index
integer
required
Required range: x >= 0
​
display_tags
string[]
required
​
prompt_index
integer
required
Required range: x >= 0
​
stream_index
integer
required
Required range: x >= 0
​
title
string
required
​
type
enum<string>
required
Available options:
title
​
summary
string

Short natural-language description of the generated track.