Run 100 ad hook variants in one Sume Format bulk run
Sume Format bulk runs take 1-100 items with concurrency 1-16. What to put in each item, and what bulk runs do not offer: no queue webhook, no cancel-queue call.
A Format bulk run takes 1-100 items and runs them with a concurrency of 1-16. Use one item per hook variant. Per Format bulk runs, there is no public cancel-queue endpoint (cancel a child with POST /v1/format-runs/{run_id}/cancel), and no queue webhook, so plan for polling.
Limits
Same key with a different body returns 409 idempotency_conflict, so derive keys from the variant content.
| Item | Value |
|---|---|
| Items per bulk run | 1-100 |
| Concurrency | 1-16 |
| Queue webhook | None |
| Cancel queue | No public endpoint; cancel one child run instead |
Item design
Keep every item identical except the hook line. That keeps the test single-variable.
Credits
If credits run short, the run can fail with insufficient_credits; check Errors and credits before launching 100.
Sources
Related posts
More in Use cases
- Edit an existing ad video with Gemini Omni Flash on Sume
Sume's Video Router takes a video_url for video edit on gemini-omni-flash-1.1. How to send an edit request and why sume/auto won't tell you the model.
- Google video experiment results: 70% directional, 95% conclusive
Google labels video experiment results directional at 70% confidence and conclusive at 95%. When to act on each, and how to line up Sume ad variants first.
- Google Ads video experiment: 2 arms or up to 10 AI variants?
Google's basic video A/B test has two arms and only covers Video Reach and Video View campaigns; Custom allows up to 10. How many Sume variants to render.
- GPT Image 2.5 icon set: one anchor icon, then reference it
Make a consistent icon set with GPT Image 2.5 on Sume: approve one anchor icon, pass it as input_references, change only the subject, and keep a style block.
Written by Sume