Video GenerationActive
Seedance v2.5 Flash Uncensored
ByteDance Seedance v2.5 Flash Uncensored creates 4–30s videos at 480p or 720p from prompts or frame images, with optional synced audio.
Text to VideoImage to Video
Model ID
seedance v2.5 flash uncensored
Provider
bytedance
Updated
Nov 16, 2026, 11:40
wiro playground—bytedance/seedance v2.5 flash uncensored
Updated Nov 16, 2026, 11:40
Overview
Seedance v2.5 Flash Uncensored is a ByteDance video generation model. It turns a written scene description into a single short video clip. You can also guide the shot with a first frame, plus an optional last frame. It outputs a ready-to-share video file, with optional synchronized audio.
What you can build
- Short social ads, promos, and product teasers
- Storyboards and pre-visualization shots for film scenes
- Image-to-video “bring this photo to life” clips
- Style-consistent clips using reference images for character, outfit, or set
- Cinematic establishing shots with specified camera movement
Inputs
- A starting frame image you upload (such as PNG, JPG, or WEBP) to drive image-to-video. The output keeps this image’s aspect ratio.
- An optional ending frame image to lock where the shot finishes. This requires a starting frame image.
- A written prompt that describes the scene, action, camera, lighting, and sound.
- 1–10 reference images to guide look and identity. In your prompt, call them out as [Image 1], [Image 2], and so on.
- A target output resolution choice: 480p or 720p.
- A target duration in whole seconds from 4 to 30.
- A choice to generate synchronized audio or return silent video.
- An aspect ratio choice. If you provide a starting frame, the model keeps that frame’s aspect ratio.
- A video container format choice (MP4 or MOV).
- An optional watermark toggle.
Outputs
- One generated video file in MP4 or MOV.
- The video matches the requested resolution and duration.
- The output keeps the selected or inherited aspect ratio.
- When audio is enabled, the file includes a synced audio track aligned to the visuals.
- When watermark is enabled, the video includes a visible watermark overlay.
Recommended settings
- Use 480p when you want fast drafts and quick prompt iteration.
- Use 720p when you want cleaner details for a clip you’ll keep.
- Start with 4–6 seconds while you test motion and staging, then increase duration.
- Use MP4 for broad playback compatibility.
- Use MOV when you plan to do color grading, keying, or compositing.
Limitations
- Flash tier output is limited to 480p and 720p.
- Maximum clip length is 30 seconds per generation.
- If you provide a starting frame image, the output preserves that image’s aspect ratio.
- Fast physical action can look inconsistent across frames.
- Real-person face references can be rejected or altered, even when prompts are allowed.
- Low-quality inputs increase failure risk. Blurry frames, heavy compression, text overlays, and existing watermarks can cause artifacts.
Safety & compliance
- Only upload content you have the rights to use.
- Don’t generate sexual content involving minors, or any exploitative sexual content.
- Don’t create non-consensual intimate content.
- Don’t use a real person’s likeness without consent.
- “Uncensored” does not mean unrestricted. The model and upstream providers can still refuse disallowed requests.
Example prompts
Great starting points for seedance v2.5 flash uncensored.
She poses for the photographer on the rooftop: she turns her shoulder, looks back into the camera, runs her hand through her blonde hair, and camera flashes go off. Golden-hour backlight, soft ocean wind, camera shutter sounds.Video Generation
She leans back against the vintage convertible and shifts between poses in the desert sun, sliding the red robe off one shoulder and smiling into the camera, palm trees swaying, warm breeze audio.Video Generation
She stays kneeling on the wooden veranda, the shoji screens behind her closed and unchanged. A soft breeze moves her long black hair and cherry blossom petals drift down around her in the dawn light. She slowly tucks her hair behind her ear, leans forward slightly with her hands on her knees, then lifts her eyes to the camera and smiles softly. Slow gentle push-in, no new objects or doors, soft birdsong audio.Video Generation
She opens the tall balcony doors, steps into the morning light, stretches and sips her coffee while a yellow tram passes on the street below, city morning ambience audio.Video Generation
API quick start
Run seedance v2.5 flash uncensored with a single API call.
POST https://api.wiro.ai/v1/Run/bytedance/seedance-v2-5-flash-uncensored
{
"prompt": "She poses for the photographer on the roo…",
"resolution": "480p",
"duration": 4,
"generateAudio": "false"
}curl
curl -X POST "https://api.wiro.ai/v1/Run/bytedance/seedance-v2-5-flash-uncensored" \
-H "Content-Type: application/json" \
-H "x-api-key: YOUR_WIRO_API_KEY" \
--data-binary @- <<'JSON'
{
"prompt": "She poses for the photographer on the roo…",
"resolution": "480p",
"duration": 4,
"generateAudio": "false"
}
JSONOther models you might like
Explore similar video generation models.