Submit 20 music takes at once: queued is normal, queue_full is not
Sume accepts music jobs as queued while the queue has room, then returns 429 queue_full. Default accepted-job capacity by plan, from 6 on Free to 120 on Scale.

Submitting twenty music takes at once works on Sume only if your workspace can accept twenty paid jobs; queued is a normal accepted state, but a full queue returns 429 queue_full. The generation admission docs, read 2026-10-03, list default accepted job capacity of 6 on Free, 24 on Pro, 48 on Startup and 120 on Scale and Enterprise, so twenty at once fits Pro and above but not Free.
What does each plan accept?
Concurrency is a dispatch limit, not a submit limit: a workspace at its processing cap still accepts jobs as queued while queue capacity remains. The docs say to prefer the effective generation_limits.concurrency_limit field over the static table, because overrides exist.
| Plan | Processing at once | Queue capacity | Accepted job capacity |
|---|---|---|---|
| Free | 1 | 5 | 6 |
| Pro | 4 | 20 | 24 |
| Startup | 8 | 40 | 48 |
| Scale | 20 | 100 | 120 |
What do I do on queue_full?
Wait for jobs to finish or cancel queued ones, then retry with the same Idempotency-Key, as the docs advise. Keep a counter of jobs you have in flight and submit the next take when one reaches completed, failed or canceled. A different error, 429 rate_limited, is abuse protection on request volume, and the docs say to back off using retry-after when present.
What does this not tell you?
- How long a queued music job waits depends on your plan's processing concurrency and on engine speed; the docs give no figure.
- Each take is billed once accepted, so queueing twenty is a $2.50 commitment at the fixed $0.125 price.
- The table is defaults. Admin overrides can raise a limit, so read your own workspace's limits.
Sources
Related posts
More in Developers
- subscribeFormatRun onCreated: save the run id, set your own key
subscribeFormatRun makes a new Idempotency-Key per call unless you pass one. Save the run id in onCreated and pass a stable key so a restart cannot double-run.
- Sume API rate limits by plan: requests per minute for writes and reads
Sume gives every API key a per-minute budget set by plan: 120 writes on Free up to 1200 on Scale, with reads at forty times the write number. Table and headers.
- Sume hides provider names and task ids: what to debug with
Sume job responses and events are provider-neutral: no vendor task ids or raw URLs. The fields to debug a video job with instead.
- Sume Auto image aspect_ratio 21:9 returns 400: what to send
sume/auto rejects aspect_ratio 21:9 and lists nine accepted values. Use image_size, or pin a model that lists 21:9, such as Nano Banana Pro or Flux 2 Pro.
Written by Sume