Create a video from a prompt
Describe one scene and a visible action. Add a camera direction, then choose a text-to-video model and its available clip settings.
AI Video Generator: From Prompts, Images or References
Describe a scene, start from a still image, or add a reference where your model supports it. Direct the action and camera, then refine in one conversation.
Ways to Create an AI Video
Animate a still image (image to video)
Attach a photo or generated image you have permission to use. Pick a model with image input and explain how the subject or camera should move.
Guide a video with reference inputs
Use supported image, video or audio references where the selected model offers them. Input roles and limits vary by model.
AI Video Examples and Prompts
Examples are creative starting points. Results depend on your prompt, model and settings; an example does not guarantee the same output.
Midnight Monster Bakery
The enormous cake suddenly rises another foot and releases a soft burst of flour. The little orange monster stumbles backward, catches the wooden spoon, blinks twice in comic disbelief, then breaks into a delighted grin. Flour curls through the warm oven light and copper pans wobble gently. Preserve the original monster design, exactly two teal horns, bakery layout and theatrical family-animation style. Smooth expressive character motion, playful timing, no new characters, no text, no logos.
Midnight Conservatory
The botanist takes two careful steps through the shallow water and raises one hand. The tiny bioluminescent seeds gather into a loose spiral above their palm, casting faint cyan reflections that move across the flooded floor. Ferns stir in a soft draft and the camera drifts slowly to the right. Preserve the greenhouse architecture, yellow raincoat, moonlit atmosphere and realistic scale. No text, no logos.
Choose a Video Model, Length and Aspect Ratio
These are examples from the integrated model catalog. Availability and supported inputs can change; the live model picker shows what you can currently use. Check the maximum credits after changing settings.
MiniMax H3
- Output options:
- 768P, 2K
- Aspect ratios:
- 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, adaptive
- Clip length in seconds:
- 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
- Supported inputs
- Text prompts · Image inputs
Seedance 2.0
- Output options:
- 480p, 720p, 1080p, 4k
- Aspect ratios:
- 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, adaptive
- Clip length in seconds:
- 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
- Supported inputs
- Text prompts · Image inputs
Veo 3.1 Fast Official
- Output options:
- 720p, 1080p, 4K
- Aspect ratios:
- 16:9, 9:16
- Clip length in seconds:
- 4, 6, 8
- Supported inputs
- Text prompts · Image inputs
Kling 3.0
- Output options:
- 720p, 1080p, 4k
- Aspect ratios:
- 16:9, 9:16, 1:1
- Clip length in seconds:
- 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
- Supported inputs
- Text prompts · Image inputs
Video Generator FAQ
Can I use a reference image?
Yes, when the selected model supports image input. Sign in to upload a reference you have permission to use. Describe which details to preserve and review the output carefully.
Which models and output settings can I choose?
The model picker lists currently available options. Image models offer their supported output sizes or formats; video models offer their supported lengths, resolutions and aspect ratios. Reference support varies by model.
How much does a generation cost?
Choose your model and settings, then read the maximum credits shown before submitting. Costs vary by request. Credits are reserved before paid work and settled after reconciliation; unused reservations are released.
How do I get a better result?
Start with a clear subject and setting. For images, describe composition and lighting; for videos, choose one movement and a camera direction. Review each result and refine one change at a time in the same conversation.
AI Image & Video Creation Guides
Plan your next prompt, add movement to an image, and understand the credits before you submit.