google/omni-flash

Gemini Omni Flash is Google's video-to-video editor. Upload an MP4 and describe an edit — recolor objects, replace props, change scenery — and Omni Flash returns the edited clip (up to 10s).

$3 / request
GPU: H100

Pricing

google/

omni-flash

Pricing for Synexa AI models works differently from other providers. Instead of being billed by time, you are billed by input and output, making pricing more predictable.

Output
$3.0000 / video
or
0 videos / $1

For example, generating 100 videos should cost around $300.00.

Check out our docs for more information about how per-request pricing works on Synexa.

ProviderPrice ($)Saving (%)
Synexa$3.0000-
replicate$5.000040.0%
fal$5.000040.0%

Readme

Gemini Omni Flash is Google's video-to-video editing model. Upload an MP4 plus a text prompt and Omni Flash returns the edited clip — recolor objects, replace elements, swap props, change scenery — while keeping the rest of the scene intact.

Capabilities

  • Text-driven edits applied to an uploaded source video.
  • Output length matches the source clip (capped at 10s).
  • Output audio is regenerated by the model (the source audio track is replaced).
  • Aspect ratio is detected automatically from the source video.

Limits and constraints

  • Source must be MP4 (video/mp4), max 100MB.
  • Output capped at 10s — longer sources are trimmed automatically.
  • Source audio is regenerated by the model. If the source audio trips Google's audio filter the job fails — Synexa pre-strips the source audio before generation to avoid this.

Tips

  • Provide a short, concrete edit instruction ("change the green bag to pink, keep everything else the same") for the most predictable results.