Sume Free plan and Omni: one running, five queued, seventh gets 429
On a Sume Free workspace one video job runs and five wait; a seventh submit returns 429 queue_full. What that means for an Omni test batch and its cost.

A Sume Free workspace processes one generation at a time and holds five more in its queue, so six accepted jobs is the most you can have in flight. A seventh valid submit returns 429 queue_full. For an Omni test run that means at most six clips outstanding: ten-second 720p clips would reserve roughly $7.60 in total. Google's Tier 1, by contrast, limits spend per 10 minutes, not job count.
Where the numbers come from
Generation admission gives processing concurrency by plan (Free 1, Pro 4, Startup 8, Scale 20) and a default queue capacity of the larger of 3 and five times concurrency. For Free that is 5 queued, so 6 accepted. Sume reserves provider list times 1.25 per the Video Router page, which puts a ten-second 720p Omni clip at about $1.27 (computed from Google's token rate; confirm in the live catalog).
| Plan | Processing | Queued | Accepted | Reserve for a full queue at about $1.27 per clip |
|---|---|---|---|---|
| Free | 1 | 5 | 6 | $7.50 |
| Pro | 4 | 20 | 24 | $30.00 |
| Startup | 8 | 40 | 48 | $60.00 |
| Scale | 20 | 100 | 120 | $150.00 |
What to do on the seventh submit
Read the error code, not just the status. queue_full means wait for jobs to finish, then retry with the same idempotency key; rate_limited means slow your submit rate; 402 insufficient_credits means top up. The retry rules are in Errors and credits.
Do not loop on queue_full without sleeping. Poll the jobs you already hold, and submit the next clip only when one finishes.
Compared with Google's limit
Google's rate limits page caps Tier 1 spend at $10 per rolling 10 minutes, so nine or ten ten-second clips at about $1.01 each would fill a window. Sume caps concurrency instead, so six $1.27 clips on Free can all be accepted at once, but they finish one after another. For a quick evaluation, the Free queue is enough; for a launch batch, check the plan ladder before the deadline.
Sources
Related posts
More in Developers
- Sume image 400s name the valid values: retry in code
Sume's Image API 400 bodies carry details.supported, allowed, min and max. A tested error table and a small function that retries with a valid value.
- Sort Sume image models by created? The field is one shared value
GET /v1/images/models returns the same created timestamp for every model, so it cannot tell you which is newest. What to use instead, with a script to prove it.
- Split a Sume job's queue wait from its run time with job events
Read GET /v1/jobs/{id}/events and subtract timestamps: job.queued to job.started is queue wait, job.started to the terminal event is run time. Python sample.
- Export every Sume job to CSV with next_cursor pagination in Python
Loop GET /v1/jobs?limit=100 and pass next_cursor back as starting_after until it is absent, then write id, status and captured cost to CSV with csv.DictWriter.
Written by Sume