gemini-omni-1.2-flash

Text to Video

Image to Video

Reference to Video

Video to Video

Model:

Generate 3–15 second videos at 360p to 4K from text, first and last frames, or up to 10 reference images and 3 reference videos, and edit existing clips up to 15 seconds with Gemini Omni 1.2 Flash API.

Get APIKey

API Doc

Input

Prompt

1287/20000

Create one continuous 15-second cinematic story at dusk as a storm rolls in. Reference image 1 is the lighthouse keeper: keep her face, silver braid, mustard-yellow oilskin coat, navy knit sweater, rubber boots and brass storm lantern exactly as shown. Reference image 2 is her border collie: keep the black-and-white coat, white face blaze, amber eyes and red neckerchief. Reference image 3 is the location: keep the red-and-white striped lighthouse, the keeper's cottage, the fenced gravel path and the granite headland. 0-5s: a wide aerial shot drifts toward the headland as dark clouds swallow the last orange light and waves explode against the rocks. 5-10s: the camera tracks low beside the keeper and the collie as they hurry up the wind-swept path, her coat snapping in the gale, the dog running ahead and glancing back at her, rain starting to streak across the frame. 10-15s: on the iron gallery at the top of the tower, she raises the lit lantern, the great lamp behind her flares to life, and its beam sweeps out across the dark sea toward a small fishing boat that turns toward the light while the collie stands beside her. Keep both characters consistent in every shot, realistic weather and water physics, natural film color. No text, no logos, no extra people or animals.

Image UrlsAdded 0/2

Drop files here, paste files, or add a URL. Optional first-frame and last-frame images, in order. Add 1 or 2 images. Using frames clears reference media and the video to edit.

Reference Image UrlsAdded 3/10

Drop files here, paste files, or add a URL. Optional reference images. Add up to 10 images; they can be combined with Reference Video Urls.

——

——

——

Reference Video Urls

Drop files here, paste files, or add a URL. Optional reference videos. Add up to 3 videos, each up to 15 seconds long.

Video Url

Drop files here, paste files, or add a URL. Optional video to edit, up to 15 seconds long. Editing uses the prompt and resolution, and the output follows the source video length.

Resolution

Duration

Aspect Ratio

Output

Example Output

This is sample data. Generate a new result to see real output.

Task ID:SLSED2I0EUQ5FWW0Created:2026-11-07T05:19:27Z

Status:finishedProgress:100%

Examples

Pricing details

Pay as you go. No subscription, no hidden fees.

1 credit = $0.005

Tier	PoYo price
360p	$0.027/sec 5.46 credits/sec 5-second video: $0.137 · 10-second video: $0.273
720p	$0.082/sec 16.38 credits/sec 5-second video: $0.409 · 10-second video: $0.819
1080p	$0.123/sec 24.58 credits/sec 5-second video: $0.615 · 10-second video: $1.23
4K	$0.246/sec 49.14 credits/sec 5-second video: $1.23 · 10-second video: $2.46

How it's billed: Billed per second of output video.

Charges are based on the final output.

Gemini Omni 1.2 Flash

Text, frames, references, and video editing in one video API

Gemini Omni 1.2 Flash API - Generate and Edit Videos up to 15 Seconds

Access Gemini Omni 1.2 Flash API instantly on PoYo. Create 3 to 15 second clips from a prompt, animate between first and last frames, combine up to 10 reference images with up to 3 reference videos, or rewrite an existing clip of up to 15 seconds with a single text instruction.

Pick any duration from 3 to 15 seconds, render at 360p, 720p, 1080p, or 4K in 16:9 or 9:16, and pay per output second through one asynchronous API with task polling and webhook callbacks.

Get API Key

Key Features of Gemini Omni 1.2 Flash API

Four production workflows in one model ID: flexible-length generation, first-and-last-frame control, multi-reference creation, and prompt-driven video editing.

Generate 3 to 15 Second Clips at Any Length

Fixed duration tiers force you to pad short shots or cut long ones. Gemini Omni 1.2 Flash API accepts any whole-second length from 3 to 15 seconds, so each clip matches its edit slot, ad format, or storyboard beat. A full 15-second clip leaves room for several shots in one take, like this four-shot macro explainer inside a beehive. Render at 360p for fast drafts and move to 720p, 1080p, or 4K for delivery.

Any duration from 3 to 15 seconds
Room for multi-shot sequences in one clip
360p, 720p, 1080p, and 4K output

Animate Between First and Last Frames

Define exactly where a shot starts and where it lands instead of hoping a prompt gets there. Send one image to animate a starting frame, or two images to move from a first frame to a last frame in order. Here an exterior view of a miniature ramen shop pushes inside to the diner at the counter in one continuous camera move, keeping the building, signage, and character consistent. With a last frame set, clips run up to 10 seconds.

One image animates a starting frame
Two images set first and last frames
One continuous move up to 10 seconds

Keep People and Products Consistent with Multiple References

Ads and product videos fall apart when a face or garment drifts between shots. Send one model photo with front and back shots of each outfit, and Gemini Omni 1.2 Flash API produces a 15-second try-on sequence that keeps the same person while the clothes change. Each request accepts up to 10 reference images and up to 3 reference videos of up to 15 seconds each, so identity, products, and motion can come from separate files.

Up to 10 reference images per request
Up to 3 reference videos, 15 seconds each
Faces and garments stay consistent across shots

Edit Existing Videos with a Text Instruction

Reshooting footage to change a setting or mood is slow and expensive. Send a clip of up to 15 seconds with a plain-language instruction, and Gemini Omni 1.2 Flash API returns an edited version that follows the original timing. Edit the result again to keep refining: this scene moves from an oak forest to a stable, a market fair, and a stone castle while the knight, the queen, and the framing stay intact.

Edit clips up to 15 seconds long
Refine the result edit by edit
Characters and framing stay intact

Who Can Benefit from Gemini Omni 1.2 Flash API?

AI video apps and social platforms

Offer users text-to-video, frame animation, and video restyling through one Gemini Omni 1.2 Flash API integration, with clip lengths that fit every feed format.

Marketing and advertising teams

Produce 6, 10, or 15 second ad cuts in 16:9 or 9:16 from product shots and brand references, then refresh existing creatives with prompt-based edits.

E-commerce and product content

Use first and last frames for controlled product reveals, and combine product photos with motion reference videos to build consistent catalog clips.

Creative studios and artists

Assemble characters, environments, and movement from up to 10 images and 3 reference videos, then iterate on style by editing the generated footage.

Previsualization and storyboarding

Turn storyboard frames into timed shots at exact lengths, draft at 360p to review pacing, and render approved scenes at 1080p or 4K.

Video workflow developers

Automate generation and editing pipelines with one model ID, asynchronous task polling, webhook callbacks, and per-second pricing that is simple to forecast.

Gemini Omni 1.2 Flash vs Gemini Omni 1.1 Flash — Model Comparison

Both models are available on PoYo. Choose Gemini Omni 1.2 Flash for longer shots, video editing, richer references, and per-second billing; choose Gemini Omni 1.1 Flash for flat per-generation prices on 8 to 10 second clips at 720p and above.

Feature	Gemini Omni 1.2 Flash	Gemini Omni 1.1 Flash
Best for	Longer shots, video editing, and multi-reference scenes	Fixed-price 4 to 10 second clips at 720p, 1080p, and 4K
Clip length	Any length from 3 to 15 seconds, so every shot fits its slot	Four preset lengths (4, 6, 8, 10s) for simple, repeatable requests
Editing existing footage	Rewrite a clip of up to 15 seconds with one text instruction	Focused on generating new clips from text, frames, and references
Reference material	Up to 10 images plus up to 3 videos of up to 15s each in one request	Up to 7 images, or one 3 to 10s reference video with up to 5 images
Generation quality	Preferred over 1.1 in Google's side-by-side tests: 63% for reference-to-video, 59% for image-to-video, 57% for editing	Established results for existing 4 to 10 second workflows
Pricing model	Per second, so short clips cost less: 3s at 720p is 49.14 credits	Per generation, with the full price known before you submit
Low-cost drafts	360p from 5.46 credits per second; 8s is 43.68 credits	One price for 360p, 720p, and 1080p: 75 credits for 8s
Long clips at 1080p and 4K	Up to 15 seconds per clip at 24.58 (1080p) or 49.14 (4K) credits per second	75 credits for 8s at 1080p and 135 credits for 8s at 4K
Reference-video jobs	Reference images and videos add no extra charge to the per-second rate	Flat 120 credits per job, or 180 credits at 4K

How to Use Gemini Omni 1.2 Flash API on PoYo

Step 1: Register & Create API Key
Sign in to PoYo and create an API key for the unified asynchronous generation endpoint.
Create API Key →

Step 2: Top Up Credits
Add credits to your account. Gemini Omni 1.2 Flash bills per output second, starting at 5.46 credits per second at 360p.
Add Credits →

Step 3: Integrate the API
Send model: "gemini-omni-1.2-flash" with a prompt. Add image_urls for frames, reference_image_urls and reference_video_urls for references, or video_url to edit a clip. Poll the returned task ID or set callback_url to receive the result.
View Documentation →

Frequently Asked Questions about Gemini Omni 1.2 Flash API

What is Gemini Omni 1.2 Flash?

Gemini Omni 1.2 Flash is a Google multimodal video model for generating and editing short videos. On PoYo, the gemini-omni-1.2-flash API creates 3 to 15 second clips from text, first and last frames, or reference images and videos, and edits existing clips of up to 15 seconds.

Which input modes does the Gemini Omni 1.2 Flash API support?

Four modes share one model ID. Send only a prompt for text-to-video, one or two image_urls for first-and-last-frame video, reference_image_urls and reference_video_urls for reference-guided video, or video_url to edit a clip. The mode is selected from the fields in each request.

How long can Gemini Omni 1.2 Flash videos be?

Choose any whole-second duration from 3 to 15 seconds, with 8 seconds as the default. When you set both a first and a last frame, durations go up to 10 seconds. Video editing returns a clip that follows the length of your source video.

How many reference images and videos can I use?

Each request accepts up to 10 reference images and up to 3 reference videos, and you can send images and videos together. Each reference video can be up to 15 seconds long. Reference media is included in the per-second price.

How does video editing work in Gemini Omni 1.2 Flash API?

Send a public video_url of up to 15 seconds with a prompt that describes the change, such as a new style, background, or mood, and optionally choose the output resolution. The edited clip follows the source timing, and billing uses the source length rounded up to whole seconds.

Which resolutions and aspect ratios are available?

Gemini Omni 1.2 Flash API outputs 360p, 720p, 1080p, or 4K, with 720p as the default. Generated clips use 16:9 landscape or 9:16 portrait framing, and edited clips keep the framing of the source video.

How much does the Gemini Omni 1.2 Flash API cost on PoYo?

Pricing is per output second: 5.46 credits at 360p, 16.38 at 720p, 24.58 at 1080p, and 49.14 at 4K, where 1 credit equals $0.005. An 8 second 720p clip costs 131.04 credits. The Playground shows the estimate before you run a task.

How do I integrate the Gemini Omni 1.2 Flash API through PoYo?

Create a PoYo API key, then POST to /api/generate/submit with model gemini-omni-1.2-flash and your input fields. Use the returned task_id to poll task status, or provide a callback_url to receive a webhook when the video is ready.

Can I try Gemini Omni 1.2 Flash before integrating it?

Yes. The Playground on this page lets you test prompts, frames, reference media, and video edits with live credit estimates, then switch to JSON mode to copy the exact request body for your integration.

Why Use PoYo for Gemini Omni 1.2 Flash API Access

One API for every mode

Generate from text, frames, or references and edit videos through the same submit and status endpoints, using a single model ID.

Transparent per-second pricing

Credits follow output seconds and resolution, and the Playground shows the estimate before each run, so costs are easy to forecast.

Playground testing

Try durations, resolutions, frames, references, and edits in the browser, then copy the exact JSON request into your code.

Production-ready callbacks

Poll tasks by ID or receive webhook callbacks when videos finish, fitting existing asynchronous pipelines without extra infrastructure.