Unlucid AIUnlucid AI image-to-video animates a still image using a motion prompt. Your image supplies the opening appearance; your words describe what moves. Upload an image you have permission to process, then choose a model in Studio.
Best for: Movement from a portrait or scene you already have.
Sign in to explore the available models and review your Credits.
Adult fashion campaign samples, created with Wan 3.0 Spicy.
Source clip: 8 s · 1280 × 720. Sample file details.
Read the campaign sample details
Campaign videos generated with Wan 3.0 Spicy from Qwen Image 2.1 images. Create your own scene in Studio; results vary with the selected model and input.
A source frame gives the video a visual starting point. It does not lock every detail: inspect the face, hands and crop as the clip develops.
No source image? Start with text
Workflow reference: Hugging Face: Image-to-Video.
Open Image to Video in Studio and sign in. Choose a model, read its upload limits and add a source image you have permission to process.
Ask for a small head turn, environmental movement or a gentle camera push. Use the image for visual detail and the prompt for action. Review output settings and the credit estimate before generating.
Watch the beginning, middle and end in History. Inspect the face, hands, background and crop against the source. If the scene drifts, reduce the movement or simplify the camera instruction for the next attempt.
Open a motion brief in Studio and add your own source image. The link fills the prompt; you choose the image and decide when to generate.
Prompt structure: What moves + how far + what the camera does
Swipe to compare the scene briefs.
Try a small movement before asking for a full performance.
The clearly adult subject makes a small, natural head turn toward the window, then holds the pose. Keep the camera steady and the lighting consistent with the starting image.
What to inspect Inspect the face at the start, middle and end. Reduce the movement if the likeness changes too much.
Use this brief in StudioKeep the subject still while testing subtle environmental motion.
Keep the main subject still. A soft breeze moves the curtain and a few loose strands of hair. Preserve the original lighting direction. A locked camera with no cuts.
What to inspect Watch the background edges and check that the request does not move the entire image as one flat layer.
Use this brief in StudioUse a gentle push-in to test framing without adding a second action.
A slow, gentle camera push toward the subject in the source image. The clearly adult subject holds the original pose. Keep the setting and light steady, with no scene change.
What to inspect Check the final crop and the subject's proportions. A camera request cannot guarantee that every source detail is preserved.
Use this brief in Studio| Before | Animate this portrait beautifully. |
|---|---|
| After | The clearly adult subject makes one small head turn toward the window. Keep the camera still and the lighting consistent with the source image. |
The image already supplies the appearance. The revision uses the prompt for movement and leaves fewer camera instructions competing with it.
A waist-up portrait supports a small turn more naturally than a full-body walk. If an image hides hands or crops the subject tightly, a large gesture asks the model to invent those details. Start with the space the image gives you.
Describe the movement you want to add: a glance toward the window, a moving curtain or a slow push-in. State what should stay steady, such as the pose or lighting. Repeating every detail of the image can obscure the action.
A source frame guides appearance but does not lock identity. Check moments when the head turns, a hand crosses the face or the camera approaches. Reduce the action if the face or proportions change in ways you cannot use.
Find more answers in Help & support.
Image-to-video content support follows the selected model's rules. The source must also meet its technical requirements, and you need the rights and adult consent for the intended use. An accepted file type alone does not make every image permitted.
The uploader accepts JPEG, PNG and WebP. Select a model in Studio to see its file-size and image-count limits before choosing a file. Technical acceptance does not establish permission to use an image.
Describe what changes from the source: subject movement, environmental motion or the camera. Start with one event. For a portrait, a small head turn with a steady camera gives you a clear movement to inspect.
The image can guide those details, but preservation is not guaranteed. Faces, hands, clothing and framing may change as the clip develops. Review the complete result rather than judging only the opening frame.
Unlucid AI
This site contains adult content. By entering, you confirm that you are at least 18 years old and of legal age to view adult content where you live.
LeaveWe’ll remember your choice in this browser for 30 days.