Help·Generate a video from an image
Generate a video from an image
Image-to-video
Bring still images to life with motion
Video generation on Endlss works by taking a source image and animating it — adding motion, camera movement, and life to a still frame. You provide the image and a prompt describing the motion you want, and the model does the rest.
All video models on Endlss are image-to-video. You'll need a source image to start. This can be any image you find while exploring, or one you've generated yourself.
How to generate a video
Step by step
1. Go to the Generate page
Navigate to Generate from the navigation menu. You'll need a Dali tier subscription or above.
2. Switch to the Video tab
At the top of the Generate page, tap the Video tab to switch from image generation to video generation.
3. Select a source image
You can set a source image by exploring Endlss and tapping "Generate like this" on any image, which will take you to the Generate page with that image pre-loaded. Alternatively, paste an image URL directly.
4. Choose your model
Pick a video model. LTX Video is the fastest and most affordable. MiniMax and Kling 2.0 produce higher quality but take longer and cost more credits.
5. Write your prompt
Describe the motion you want. For example: "gentle camera pan to the right, soft wind blowing through hair" or "slow zoom in, particles floating". The prompt guides the motion, not the content — the content comes from your source image.
6. Set duration and aspect ratio
Choose the video duration (typically 5 seconds) and aspect ratio (16
, 9, etc.) to suit your needs.7. Generate
Hit Generate. Your video will appear in your queue and be delivered to your Inbox once complete. Generation typically takes 20 seconds to 90 seconds depending on the model.
Tips for better results
Get the most from video generation
Use clear, simple motion descriptions. Models respond best to straightforward direction — "slow zoom in" works better than a lengthy paragraph. High-quality source images with good lighting and composition tend to produce better videos.
What are AI models?
Compare video models and understand which one to choose.
Generate an image from text
Create a source image from a text prompt, then turn it into a video.