8 AI music tracks for an eight-scene film score: $1.00 on Sume
Sume Music 1.0 charges a fixed $0.125 per accepted generation, so 8 tracks for an eight-scene film score is $1.00 no matter how long the prompts are.

8 tracks for an eight-scene film score cost $1.00 on Sume Music 1.0: a fixed $0.125 for each accepted generation.
Sume's music docs (read 2026-10-06) say the price does not change with prompt length or optional image conditioning. The rate card prices it as the fal list price of $0.10 per audio (Lyria 3.5) at the standard margin.
How does the cost scale with retakes?
A rejected track is a new generation, so budget for retakes.
| Retakes per keeper | Generations | Total |
|---|---|---|
| 1 | 8 | $1.00 |
| 2 | 16 | $2.00 |
| 3 | 24 | $3.00 |
What should I check?
Music is generated asynchronously like other jobs; poll or use a webhook.
- The price is per accepted generation, not per second of audio.
- Pin the prompt tempo and mood so retakes converge.
- Confirm the endpoint and limits in the docs before a batch.
Sources
Related posts
More in Pricing
- Magnific credit price: $0.66 to $1.00 per 1,000 by plan
What a Magnific credit is worth: plan price over yearly credits gives $0.66 to $1.00 per 1,000, annual billing cheapest. Derived from its pricing page.
- MAI-Voice-2.1 vs Sume text to speech: cost per 1,000 characters
Sume TTS costs 4.75 cents per 1,000 characters before the per-job round-up. A tracker lists MAI-Voice-2.1 at $0.022 and Flash at $0.015, cheaper per character.
- Manus agent calling paid Sume tools: cap spend per call
A Manus agent with a Sume API key can start paid generation. Tell it to send max_spend_usd, dry_run and an idempotency_key on every paid call. Why each matters.
- 300 images for a marketplace listings: GPT Image 2.5 cost by quality
300 1024x1024 images on Sume with ChatGPT Image 2.5: $7.4250 at low, $16.6875 at medium, $66.7125 at high quality.
Written by Sume