Brand setup for AI ad batches: a saved Format as the brand kit
Higgsfield Ads Studio starts with brand setup. On Sume, a saved Format holds your house style and output contract so each ad run starts from it.
Brand setup in an ad tool means the brand's look and rules are stored once and applied to every ad. Higgsfield's changelog lists brand setup as a step in Ads Studio. On Sume the closest feature is a Format: a saved production recipe with a house style, an output contract, and a playbook that your backend calls by name. It stores the rules as instructions, not as a visual brand-asset library.
Sume facts come from the Format API and Format catalog docs; the Higgsfield line is from its changelog, read 2026-10-02.
What does a Format store?
The Format docs describe it as a saved recipe: a house style, an output contract, and a playbook for one kind of video. A run starts in a fresh sandbox with the generation tools, and returns media on media.sume.com plus, if you ask, JSON in a schema you define. You call it at POST /v1/formats/{handle}/{slug}/runs.
Where do brand details like logo and colors go?
Put durable rules (tone, banned phrases, color words, layout habits) in the Format. Put per-ad data (product name, photo URL, price) in each run's input. The run body's input is a free-form object by design, so you choose the field names and tell the Format what they mean.
| Data | Put it in | Changes per ad? |
|---|---|---|
| Tone and style rules | The Format | No |
| Product name and price | Run input | Yes |
| Product photo | Run input or an attachment | Yes |
| Spend limit | generation_spend_cap_usd on the run | Per run |
Is there a ready-made ad Format?
The catalog lists callable Formats such as sume-product-commercial and sume-recreate, addressed at the sume handle. They are video recipes, not a static-ad generator, and the catalog page says to read one with GET /v1/formats/sume/{slug} before calling it. If you want your own brand rules, fork one in the Format library and call your copy as {your_handle}/{slug}.
What should I check before running a batch?
Run one ad, read the receipt, and fix the Format's wording before queuing more. Queue 20 ad runs overnight with Format bulk runs shows the batch call.
Write the brand rules as short, testable sentences, such as which words never appear and which color words anchor the look, instead of a long mood paragraph. A rule you can check on a finished ad is a rule the Format can follow. When an ad breaks a rule, edit the Format once, and every later run inherits the fix; that is the main reason to store rules there rather than in each request.
Sources
Related posts
More in Formats
- Cancel a Format run: cancel_effect canceled vs no_op
POST /v1/format-runs/{run_id}/cancel is idempotent and returns the current receipt. Read cancel_effect to know if you stopped it, and what still bills.
- Claude 'Schema is too complex' 400: strict tool limits vs Sume
Claude caps strict tools at 20, optional parameters at 24 and union parameters at 16 per request. Sume checks 10 levels and 5000 properties at submit.
- Claude structured output refusal or max_tokens: what Sume does
Claude can return a 200 that does not match your schema on refusal or max_tokens. Sume reports a failed projection as output_error on the receipt instead.
- POST /v1/formats: skill_slug_taken, skill_slug_reserved, 503
Creating a Format over the API can return 409 for a taken or reserved slug and 503 format_git_unavailable. What each means, and what to keep from the reply.
Written by Sume