Sound-on Reels product spot in one call: Omni Flash, 8 s, $1.00
Reels default to sound on. Omni Flash 1.1 returns synced audio: 8 s at 720p is $1.00 on Sume, against $3.27 for a silent clip plus voice and music.

Gemini Omni Flash 1.1 on Sume generates synced audio with every clip and has no switch to turn it off, so a sound-on Reels product spot is one call: 8 seconds at 720p costs $1.00 (read 2026-10-05, billed at $0.125 a second), or $1.50 at 1080p. Meta says Reels default to sound on and that 9:16 video with audio and key messages in the safe zone gets 2x higher delivery to Reels, so a clip that arrives with its own sound saves a stitching step.
What Meta says about sound
Meta's Reels ads page says 9:16 video with audio and key messages in the safe zone gave 34.5% lower cost per result than image ads and 2x higher delivery to Reels. A Meta business news post from Oct 10, 2023 reported that music plus voice-over gave a 15-point higher positive response score, and that vertical sound gave 4.8% lower CPA, 5.1% higher CTR and 2.9% higher conversion. These are Meta's figures from its own tests, not Sume's; treat them as the reason to test sound, not a promise.
What the Omni row gives you
The Sume catalog lists Gemini Omni Flash 1.1 at 3 to 10 s, 16:9 or 9:16, resolutions from 360p to 4K, with image-to-video and reference images up to 10 (videos docs). Native synced audio is always produced; there is no generate_audio toggle and no reference audio. Both the audio and the picture come from one prompt.
| Route | Lines | Total |
|---|---|---|
| Omni Flash 720p, native audio | 8 s at $0.125 a second | $1.00 |
| Omni Flash 1080p, native audio | 8 s at $0.1875 a second | $1.50 |
| Seedance 2 720p silent, plus TTS, music and render | $3.024 + $0.019 (400 chars) + $0.125 + $0.10 | $3.268 |
When to choose the longer route
The three-part route is dearer but gives control: your own script read by TTS at $0.0475 per 1,000 characters, a Lyria track at $0.125, and a Timeline render at $0.10 with ducking. Pick it when the voice must say exact words, such as a price or a code. Omni audio comes from the prompt and cannot be edited into a script.
If you want music under the Omni audio, detach and duck it, as in the Omni music bed recipe. Clips are capped at 10 s, so an ad longer than that needs two clips joined in Timeline.
- Keep the product name and any price out of the audio prompt; add them as caption cues instead.
- Review audio before captions: a clip with bad sound is a reject even if the picture is fine.
- Use 720p for the test and move winners to 1080p.
For the Reels safe-zone brief that goes with this clip, see the Reels sound-on and safe-zone brief, and for Omni limits and 4K pricing see the Omni Flash length and price note.
Sources
Related posts
More in Models
- Spain vs Mexico, Brazil vs Portugal: regional voiceover on MAI, Sume
MAI-Voice-2.1 lists es-ES, es-MX, pt-BR and pt-PT. Sume's guard compares only the primary language. How to pick a regional read and test it for 1 cent.
- Text edits on Sume: use sume/auto or pin ideogram/ideogram-v4.5?
Sume's sume/auto never says which model ran, so a text-edit workflow that depends on Ideogram 4.5 behavior should pin ideogram/ideogram-v4.5.
- Sume image models that do not list the 4:5 Instagram ratio
Imagen 4, Grok Image and Higgsfield Soul do not list 4:5. Which Sume ids do, what 4:5 means in pixels, and the nearest ratio to use when a model lacks it.
- Summer 2026 video model features mapped to Sume routes, and the gaps
Seedance 2.5, Wan 3.0, Omni 1.1 Flash and Ray3.2 launched features this summer. Which map to a Sume route, which are vendor-only. Read 2026-10-05.
Written by Sume