12 game music tracks on lyria-3.5: $1.50 with a flat $0.125 each
A 12-track game soundtrack costs $1.50 on the Sume music router: $0.125 per generation regardless of track length. Write the length into the prompt.

Twelve game music tracks cost $1.50 on Sume's music router, because every generation is a flat $0.125 whatever the track length. You cannot send a duration field, so you ask for the length inside the prompt.
Cost of a soundtrack set
Flat price per generation, multiplied by the number of tracks you keep or discard. Rerolls cost the same as keepers.
| Generations | Rate | Total |
|---|---|---|
| 6 | $0.125 | $0.75 |
| 12 | $0.125 | $1.50 |
| 24 | $0.125 | $3.00 |
| 36 | $0.125 | $4.50 |
One prompt per track
Put the tempo, key, instruments and length in the prompt, plus the exclusions (the negative_prompt field must be empty). For a loop, ask for a seamless 30-second loop; Sume does not enforce the length, so measure the file afterward.
Use a different Idempotency-Key per track so a retry cannot duplicate one.
curl -X POST https://api.sume.com/v1/music-router/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: game-forest-day-01" \
-d '{"model": "sume/music-auto",
"prompt": "Gentle forest village theme, 92 BPM, D major. Plucked harp, soft flute. A 30-second seamless loop. Instrumental, no vocals, no drums.",
"metadata": {"level": "forest", "slot": "day"}}'Limits
The prompt is 1 to 5,000 characters. duration and duration_seconds are rejected. sume/music-auto currently routes to lyria-3.5, and job.request.routed_model reports which engine ran. Pin lyria-3.5 explicitly if you want the same engine next month, since the auto alias may route elsewhere later.
Sources
Related posts
More in Use cases
- 12 team avatars plus one 20-second clip each: $11.40 + $58.80 = $70.20
12 team avatars cost 12 x $0.95 = $11.40 on Sume; one 20-second plus clip each adds 12 x $4.90 = $58.80. Total $70.20, or $55.56 at standard.
- 12 product photos to 4-second clips: price by resolution
Twelve product photos animated into 4-second clips cost $6.00 on Wan 3.0 at 720p. Totals at 480p, 720p and 1080p on Wan 3.0, Omni Flash and Seedance 2.5.
- 14-page read-aloud video: per-page TTS, one concat, page-turn offsets
Narrate 14 pages of 160 characters with Sume TTS ($0.11), join for $0.01, set page starts from segments[]. About $0.32 with a 2-minute render.
- 15 thumbnail tests (5 videos x 3 variants): $2.25 at 2K on Sume
Three 16:9 thumbnail variants for each of five videos is 15 images: $1.50 at 1K, $2.25 at 2K or $3.00 at 4K on Nano Banana 2.1 via Sume. YouTube's size rules.
Written by Sume