Music from a still: image_url conditioning costs the same $0.125

Yes: Sume Music can take an image alongside the prompt. Send an optional image_url that is a public HTTPS image, and the model uses it as conditioning. The price does not change with the image or the prompt length; the docs set a fixed $0.125 per accepted Music 1.0 generation. The prompt is still required and stays the main control.

Rules for the image input

Pairing the still with a brief

The docs advise passing the accepted scene still as image_url when it applies. Keep the written brief concrete anyway: emotion, genre, tempo as a number, key, two to four instruments, an arc with one named moment, and an era. The image shapes mood; it does not replace those choices.

For scenes in one project that should contrast, the docs say to change the broad genre family, move the tempo at least 12 BPM apart, and change the lead instrument. For one consistent score, keep those fixed and reuse the same brief.

Request

The docs' own image-conditioned example.

curl -X POST https://api.sume.com/v1/music-1.0/generate \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: music-image-001" \
  -d '{
    "prompt": "Cinematic ambient underscore matching the mood of the reference still, instrumental only",
    "image_url": "https://example.com/moodboard.png"
  }'

Cost for a storyboard

Music is per scene if you want each scene scored. Six scenes with one still each is 6 x $0.125 = $0.75, the same as six text-only generations. For the router route (POST /v1/music-router/generate) the price is the same fixed Music price, so the choice of endpoint does not change the bill.

Common mistakes

Do not send a signed, private or localhost URL; the image must be public HTTPS. Do not expect the image to override the prompt. If the track needs to be vocal-free, say so in the prompt: "Instrumental, no vocals."

Use Sume media URLs from the result when you pass the audio on. The docs say raw provider URLs are not public outputs.