Check every Meta ad crop for cut-off text with video frames
With ad-set placement exclusions removed, each crop can run anywhere. Pull stills from every cut with Sume video frames and check the price and headline.

Pull three stills from every crop of an ad and read them before you launch: the first second, the middle and the last frame. Sume's video frames route returns durable images at the times you name and is unbilled, so checking costs nothing but a minute.
The reason it matters now: the Common Thread summary says Meta removed ad-set placement exclusions on 2026-08-21 (read 2026-10-02). When you cannot opt out of a placement, a crop that hides the price is a live risk.
How does video frames work?
POST /v1/video-frames takes one workspace media.sume.com clip and either at[] (1 to 24 times in seconds) or fps, and returns image artifacts. The source must be 300 seconds or shorter. Import an off-host file first with POST /v1/media-imports.
Submit is always 202; poll GET /v1/video-frames/{id} until resource_status is ready, then read frames[{t,url,width,height}]. If one instant fails, that frame's url is null and the job does not fail.
Which frames should I look at?
Three per cut is enough for a first pass. Use the first second because that is where the hook text appears, a middle frame for the product, and a time near the end for the price or call to action.
| Check | Time (`at`) | What to look for |
|---|---|---|
| Hook | 0.5 | Headline fully inside the frame |
| Product | Middle of the clip | Product not cropped at the edge |
| Call to action | Last second | Price and button text readable |
What do I set for the image size?
max_edge accepts 16 to 2160 and clamps the long edge. Omit it to keep the source size. Use format: "png" when you need lossless stills for a close read of small type.
What if a crop fails the check?
Re-render that crop rather than the whole set. A Timeline 1.0 render can set output size and fit per cut, and it costs $0.10 per output minute rounded up, per the docs; confirm the live rate in GET /v1/catalog.
Sources
Related posts
More in Media tools
- Add captions to a MiniMax H3 clip: video-captions, silent clips
A minimax-h3 clip carries native stereo audio, so video-captions can transcribe it. Silent clips fail with caption_no_speech; pass cues instead.
- Pull the audio out of a MiniMax H3 clip with Sume audio-detach
A minimax-h3 clip has native stereo audio. audio-detach returns it as a wav or mp3 artifact without touching the video. Fields, defaults and caps.
- MiniMax H3 reference formats: HEIC, MOV, WAV, MP3 limits
MiniMax H3 accepts HEIC/HEIF images, MP4/MOV video and WAV/MP3 audio as references. A pre-flight checklist before you submit minimax-h3 on Sume.
- Keep the sound of AI clips in a Sume timeline: detach to a spine
A Timeline 1.0 render takes one audio spine. To keep audio generated with your clips, detach each clip's track, join the parts, and use that as the spine.
Written by Sume