OmniaKey Studio
AI Video Generator
Give your ideas a little motion.
From a moment to a moving story
Start with your own idea or make this prompt yours.
Uses your OmniaKey balance at the standard model rates.
View pricingHow this generator works
- 01
Sign in and choose an operation
Sign in and pick text-to-video, image-to-video or reference-to-video. Each operation expects different inputs.
- 02
Write the shot
Describe one shot: subject, action, camera and setting. The model turns that description into motion, not a full screenplay.
- 03
Submit and let it run
Video tasks run in the background. You can leave the page; progress stays available under My creations.
- 04
Review, then download
Results are kept for 30 days. A failed task produces no result and returns the reserved credit.
Models in this generator
11 video models are listed here. Prices come from the same catalog the workbench uses.
| Model | Provider | Operations | Listed price |
|---|---|---|---|
| Seedance 2.0 Fastdreamina-seedance-2-0-fast-260128 | ByteDance | Text to video · Image to video · Reference to video | $0.03–$0.06 / second |
| Seedance 2.0 Minidreamina-seedance-2-0-mini-260615 | ByteDance | Text to video · Image to video · Reference to video | $0.02–$0.04 / second |
| MiniMax H3minimax-h3 | MiniMax | Text to video · Image to video · Reference to video | $0.0269–$0.0431 / second |
| HappyHorse 1.1happyhorse-1.1 | Alibaba | Text to video · Image to video · Reference to video | $0.042–$0.108 / second |
| HappyHorse 1.0happyhorse-1.0 | Alibaba | Text to video · Image to video · Reference to video | $0.112–$0.192 / second |
| Wan 3.0wan3.0-video | Alibaba | Text to video · Image to video · Reference to video | $0.04–$0.16 / second |
| Seedance 2.5dreamina-seedance-2-5-260628 | ByteDance | Text to video · Image to video · Reference to video | $0.1133–$0.6259 / second |
| Seedance 2.0dreamina-seedance-2-0-260128 | ByteDance | Text to video · Image to video · Reference to video | $0.077–$0.407 / second |
| Wan 3.0 Video Primewan3.0-video-prime | Alibaba | Text to video · Image to video · Reference to video | $0.078–$0.38 / second |
| Grok Imagine Video 1.5grok-imagine-video-1.5 | xAI | Text to video · Image to video | $0.30 / second |
| Kling 3.0kling-3.0 | Kling | Text to video · Image to video | $0.80–$3.00 / request |
Prices come from the model catalog. Your balance is charged at the model's standard rate; check the pricing page for the current table.
Prompt guidance
One shot per prompt
A single camera move and a single action are easier to follow than a sequence of scenes.
A cyclist turns a corner on a wet street, camera tracking from the side
Describe the action
Say what moves, in which direction, and how fast. Static subjects leave the model guessing.
Hair moves in the wind; the paper lifts off the table
Direct the camera
Name the camera move and framing. Keep them compatible with the action.
Slow push-in, medium shot, eye level
Respect clip length
Short clips suit a single beat. Plan cuts in your editor instead of asking one generation for a whole sequence.
Five-second loop of steam rising from a cup
Use a reference image deliberately
A reference fixes the subject or first frame. Describe only what should change, so the model does not redesign the reference.
Keep the product and background; add rain and a slow dolly
More on these models
These articles take a closer look at the video models and workflows mentioned above.
- Wan 3.0 Prompt Guide: Structure, Examples and Real Clips
- Hailuo API: MiniMax H3 Python Guide and 3 Video Modes Tested
- HappyHorse 1.1 vs Kling 3.0: Same-Prompt Test and Price
- HappyHorse 1.1 vs Seedance 2.0: Same-Prompt Test and Price
Common questions
Which video operations are available?
The generator lists text-to-video, image-to-video and reference-to-video models it can run. Available operations and models depend on your account.
How long does a video take?
Generation time depends on the model, resolution and clip length. The task runs in the background, so you can leave the page and return later.
Can I generate people or branded content?
Use only material you have the rights to. The model provider's terms and your local law apply to likeness, trademarks and generated output.
What happens if a video task fails?
A failed task produces no result and returns the reserved credit. Start a new task with a revised prompt instead of resubmitting the same one blindly.
Can I use these models through the API?
Yes. The same models are available through the OmniaKey API with the same API key. See the model catalog for model pages and the API documentation for request formats.