Gemini Omni Flash 1.2 Image-to-Video
Gemini Omni 1.2 Flash image-to-video: the image is the first frame, an optional End Frame is the last (10 seconds max with an end frame, 15 without), native audio
Gemini Omni Flash 1.2 Image-to-Video
Node type: video:gemini_omni_flash_12_i2v
Category: video
Hidden in sidebar: yes (auto-selected by routing)
Description
Gemini Omni 1.2 Flash image-to-video: the image is the first frame, an optional End Frame is the last (10 seconds max with an end frame, 15 without), native audio
Pricing
- Cost: ~14 credits per second
- Per-second rate: The figure is the rate for a 5-second clip. Shorter clips can cost a bit more per second. You're always charged for the real length.
Variant family
- Family:
gemini-omni-flash-12 - Is primary: no
- All variants:
video:gemini_omni_flash_12_t2v,video:gemini_omni_flash_12_i2v,video:gemini_omni_flash_12_r2v,video:gemini_omni_flash_12_edit
Canvas ports
These appear as port handles on the left side of the node.
| ID | Label | Details |
|---|---|---|
image_url |
Image (start frame) | IMAGE (required) |
end_image_url |
End Frame (max 10s) | IMAGE |
prompt |
Prompt | TEXT (required) · multiline |
Sidebar config
These render as form fields in the right-side config panel when the node is selected.
| ID | Label | Details |
|---|---|---|
duration |
Duration (seconds) | NUMBER · 3…15 · default: 8 |
aspect_ratio |
Aspect Ratio | TEXT · options: 16:9, 9:16 · default: "16:9" |
resolution |
Resolution (up to 4K) | TEXT · options: 360p, 720p, 1080p, 4k · default: "720p" |
Outputs
| ID | Label | Type |
|---|---|---|
video |
Generated Video | VIDEO |
How to prompt Gemini Omni 1.2 Flash
- One take or several. It cuts between shots by default. For a single take, write "single continuous shot, no scene cuts".
- Multi-shot timing. Give each shot its own time range, for example "[0-4s] a wide shot of the street, [4-8s] close on her face".
- Dialogue. One speaker per clip. Quote the exact words, keep the lines short, and say who speaks and how it sounds: "the old man says quietly, 'Not today.'" For a conversation, make one clip per speaker and cut between them in the editor.
- Reference images. Up to 6 in total, up to 3 per character. Describe what each image is for, in the order you wire them: "the first and second images are the same woman, the third is her dog".
- Edits. One short instruction, then "Keep everything else the same", for example "make it night. Keep everything else the same." Every edit sends the clip as a file, so expect video edits to be unavailable in the EEA, the UK and Switzerland, for clips Omni made as well as your own footage.
- First and last frame. Wire a start image and an end frame and describe the move between them. Runs with both frames are 10 seconds at most.
- Resolution. Draft at 360p and keep the take you like. 1080p and 4K make a NEW take from the same prompt, not an upscale of the draft, so the result will differ.
- Sound. There is no audio input. Describe the sound and music in the prompt ("Sound design: rain on a tin roof, no music"), and add "No dialogue" when nobody should speak.
Auto-generated from the Wireflow node registry.