Lands on your frames
Give it a start image, or a start and an end image, and the clip opens on the first and ends on the second, in the motion you describe.
The Gemini Omni Flash API: Google's video model turns text, a start frame, a start and end frame or up to 3 reference images into 720p or 1080p video with sound, from $0.047 per second.
gemini-omni-flashimport os, time, requestsAPI = "https://api.zurelay.com/v1"headers = {"Authorization": f"Bearer {os.environ['ZURELAY_API_KEY']}"}res = requests.post(f"{API}/videos", headers=headers, json={ "model": "gemini-omni-flash", "prompt": "A red fox trotting through fresh snow at dawn, cinematic", "seconds": 5, "resolution": "720p", "aspect_ratio": "16:9",})res.raise_for_status()job = res.json()# Takes a few minutes: poll until it's donewhile job["status"] in ("queued", "in_progress"): time.sleep(5) job = requests.get(f"{API}/videos/{job['id']}", headers=headers).json()if job["status"] == "failed": raise SystemExit(job["error"]["message"])video = requests.get(f"{API}/videos/{job['id']}/content", headers=headers)with open("video.mp4", "wb") as f: f.write(video.content)print(f"Saved video.mp4 for ${job['cost']}")Base URL https://api.zurelay.com/v1. Plain JSON over HTTPS: start a job, poll it, download the MP4. Omni Flash guide
Pricing
Pay as you go from prepaid credit, with no subscription. Top up from $10. A video is charged only once it's delivered; videos that fail cost nothing.
A clip costs its length in seconds times the price at its resolution. It's held from your balance when the job is accepted and charged when the video is delivered. First and last frames, reference images and the aspect ratio don't change the price.
Pick a resolution and a length, and how many videos you make.
100 clips × $0.282
Overview
Gemini Omni Flash is Google's video generation model, built to understand text, images, audio and video together and to make video with its sound in one pass. On Zurelay it makes 720p and 1080p clips of 6 or 8 seconds from a prompt, a start frame, a start and end frame, or up to 3 reference images. The Zurelay Gemini Omni Flash API costs $0.047 per second of video, and a video is charged only once it's delivered.
Google describes Gemini Omni Flash as a high-performance multimodal model for high-speed video generation, editing and cinematic control. It processes text, images, audio and video together, and makes video with audio from a text description. Google released it in the Gemini API as a preview on June 30, 2026. On Zurelay it uses the model ID gemini-omni-flash.
On Zurelay, Gemini Omni Flash makes video from a text prompt, from a start frame, from a start and an end frame it moves between, or from up to 3 reference images of people, products or places you describe in the prompt. Sound is always on: describe the dialogue, effects and music you want in the prompt. Clips run 6 or 8 seconds, at 720p or 1080p, in 16:9 or 9:16.
Video takes time, so the API is asynchronous. POST https://api.zurelay.com/v1/videos returns a job right away; poll GET /v1/videos/{id} until its status is completed, then download the MP4 from /v1/videos/{id}/content or from the signed url in the job. A clip usually takes about 1.5 minutes. Videos are kept for 30 days.
Zurelay prices Gemini Omni Flash by the second, the same at both resolutions: $0.047 per second, so a 6 second clip costs $0.282. Google lists the model at about $0.10 per second at 720p and $0.15 at 1080p. Frames, reference images and the aspect ratio don't change the price. The price is held from your balance when the job is accepted and charged only when the video is delivered; videos that fail, including ones the content filter refuses, are free.
Strengths
Give it a start image, or a start and an end image, and the clip opens on the first and ends on the second, in the motion you describe.
Dialogue, effects and ambience come with the picture. Describe the sound you want in the prompt, and time it to the action.
Zurelay charges the same per second at 720p and 1080p, so the sharper picture costs nothing extra.
Google's guide shows prompts with timecodes such as [0-3s] and [3-6s] to place each beat of the action within the clip.
Use cases
Start from a product shot, a portrait or a landscape and describe the camera move and the sound.
Give a start and an end frame, such as day and night or empty and furnished, and let the clip move between them.
Pass a product photo as a reference image and describe the scene to put it in.
Make 9:16 clips for stories and shorts, with sound, ready to post.
Get started
No waitlist and nothing to install: three HTTPS endpoints and your API key.
Sign up, add credit and create an API key. Set a monthly budget or a rate limit per key if you like.
Every video endpoint lives under this base URL. Send your key as a Bearer token.
https://api.zurelay.com/v1POST a prompt to /v1/videos, poll the job every few seconds until it's completed, then download the MP4. Videos are kept for 30 days.
gemini-omni-flashWorks with the tools you already use
Compare
from $0.071 per second
Pick Grok Imagine Video 1.5 for clips from 1 to 15 seconds, with sound you can turn off.
Grok Imagine Video 1.5 APIfrom $0.03 per second
Pick Vidu Q3 Turbo for clips up to 16 seconds at 1080p, from a first to a last frame.
Vidu Q3 Turbo APIfrom $0.05 per second
Pick Wan 3.0 for up to 10 reference images and clips up to 30 seconds.
Wan 3.0 APIZurelay charges $0.047 per second of video at 720p and at 1080p, so a 6 second clip is $0.282. Google lists the model at about $0.10 per second at 720p and $0.15 at 1080p. Frames, reference images and the aspect ratio don't change the price, and a video is charged only once it's delivered.
Send POST https://api.zurelay.com/v1/videos with model gemini-omni-flash, a prompt describing the motion and sound, and first_frame set to an image (an https link or a base64 data URL). Add last_frame to end on a second image. Set aspect_ratio to the image's shape, 16:9 or 9:16.
6 or 8 seconds on Zurelay. Leave out seconds and you get 6. Each clip is generated in one pass with its sound.
720p and 1080p, in 16:9 or 9:16. Both resolutions cost the same per second. 360p and 4K aren't available on Zurelay yet.
Yes, up to 3 in reference_images, as https links or base64 data URLs. Describe in the prompt who or what from each image appears and what happens. Reference images aren't used as literal frames; to start or end on an image, use first_frame and last_frame instead.
When Zurelay accepts the job, its price is held from your balance, so videos in progress can never spend more than you have. It's charged only when the video is delivered. If a video fails, including when the content filter refuses the prompt or images, it's free.
Not yet. Google's own API can edit and extend videos in a conversation; on Zurelay, Gemini Omni Flash makes new clips from text and images.
Google says every video from the model carries SynthID, an invisible watermark that identifies it as AI-generated. Zurelay is an independent service and is not affiliated with or endorsed by Google.
Create a key in seconds, point the SDK you already use at Zurelay, and every request costs up to 90% less from the first token.