
Seedance 2.5 Flash Uncensored
Seedance 2.5 Flash Uncensored is ByteDance's video model, hosted on Vellria with nothing to install: text to video and image to video, 480p to 720p, clips of 4–30s with sound. Run it in Studio or over the REST API, from 35 credits / s (100 credits = $1.00).
vellria-seedance-2-5-flashSeedance 2.5 Flash is developed by ByteDance. Vellria provides access to the model; it does not own it. All trademarks belong to their respective owners.
Output
No generation yet
Cover art
from 35 credits / s
Your request will cost 175 credits.
About Seedance 2.5 Flash
Also known as: Seedance Flash · ByteDance Seedance 2.5 Flash
Seedance 2.5 Flash is the Flash edition of Seedance 2.5, the newest video model from ByteDance's Seed team, the group that also makes Seedream. It is a sibling rather than a cut-down copy: the same family habits, the same take lengths and the same generated sound, tuned to come back sooner. The trade is resolution. Flash stops at 720p, with 480p below it for drafts, and the 1080p step stays on the full model. If you were looking for a Seedance NSFW generator without a content filter and do not need a full-HD master, this is the quick way in.
There are three ways to start a clip. Write the scene from text and pick the frame shape, or leave it on adaptive and let the model choose. Upload a still and it becomes the first frame; the clip keeps that picture's proportions, so crop it before uploading. Or set the reference mode to Character reference and attach up to ten pictures of the people, outfits or rooms that should appear. The prompt cites them as @image1, @image2 and so on, in the order you attached them, and the model builds a fresh scene around them instead of animating one fixed frame.
Sound is rendered together with the picture and follows what you write: footsteps on tile, rain against glass, a slow song from the next room. Audio is on by default; switch it off in the controls when you plan to lay your own track. Takes can run long without a cut, so describe the action as one continuous sequence in the order it happens. A long take holds together better as a single movement than as a list of unrelated beats crammed into one clip.
At 480p and 720p, Flash is priced per second exactly like Seedance 2.5, so the choice between the two is not about budget: Flash for quick turnarounds and drafts, the full model when the keeper has to be 1080p. Length still raises the price, so test staging on short takes and spend on the long one once it works. For shorter scenes at a lower price per second, Seedance 2.0 is the earlier generation, with the same start-frame workflow.
Seedance 2.5 Flash in action
Seedance 2.5 Flash outputs with the exact prompts behind them.
Image to video
She keeps walking across the station hall toward the light, her trench coat swaying with each step, dust drifting in the sunbeams, smooth tracking shot
Image to video
She stretches on the sun lounger, pushes the sunglasses back into her hair and turns her head toward the camera with a relaxed smile, palm shadows swaying in the breeze, slow push-in
Text to video
A woman in her thirties in an ivory satin slip dress walks barefoot along a moonlit beach as the tide washes over her feet, the warm wind lifting her hair, smooth tracking shot
When to pick Seedance 2.5 Flash
Pick it when
- You want Seedance 2.5 scenes with generated sound and 720p is enough for where the clip will be shown.
- You are drafting staging, pacing and lines and want each take back sooner before deciding which one to keep.
- You have pictures of the people, outfits or setting and want a new scene built around them rather than one fixed frame animated.
Pick another model when
- The final clip has to be 1080p; Seedance 2.5 offers the same three modes with a higher resolution step.
- Your scene fits a shorter take and you want to pay less per second; Seedance 2.0 is the earlier generation built for that.
- You want a long single take with a different feel to the motion; run the same prompt on Wan 3.0 and compare.
Prompt or reference
Describe the shot in words when it is new and pick its frame shape, or leave the shape on adaptive. Upload one still when the opening image is decided; it starts the clip and sets the proportions. Attach several pictures as character references when people, outfits or rooms must reappear, and point to each one from the prompt.
How to use Seedance 2.5 Flash on Vellria
- Prompt: up to 10,000 characters in the panel above.
- Cost: from 35 credits / s — see pricing.
Example prompts
An adult woman in a linen shirt opens the curtains of a seaside hotel room at dawn, turns back toward the bed and smiles, waves and gulls outside, the camera holds steady at the doorway.
From the uploaded still: the adult man leans back in the leather armchair, loosens his tie and lifts his glass toward the camera; low crackle of a fireplace.
Character reference: the adult woman from @image1 and the adult man from @image2 slow-dance in the dim lounge from @image3, warm amber light, a saxophone playing, the camera circles them once.
An adult couple share an outdoor shower on a jungle villa terrace, steam rising in the evening air, cicadas and running water, single continuous take.
Vertical frame: an adult model in a satin robe walks down a neon-lit corridor toward the camera, heels clicking on concrete, muffled club music behind the walls.
Tips
- Draft at 480p to settle staging and timing, then run the same prompt at 720p for the version you keep.
- In Character reference mode, give every picture a job in the prompt, such as the face from @image1, the outfit from @image2 and the room from @image3, and cite only images you attached.
- Name the sounds you want, such as rain, breathing or a song, or switch audio off when you will score the clip yourself.
- For a start frame, crop the still to the shape the clip should have; the frame-shape menu applies to text and Character reference runs.
Specifications
| Resolution: | 480p720p |
|---|---|
| Duration: | 4–30s |
| Aspect ratio (Text to video): | Auto (chosen from prompt)21:916:94:31:13:49:16 |
| Reference images: | 1 reference image |
See also: uncensored AI video generator · NSFW video generation · NSFW image to video.
Pricing
| Resolution | Credits |
|---|---|
| 480p | 35 credits / s |
| 720p | 75 credits / s |
Seedance 2.5 Flash vs Seedance 2.5 and Seedance 2.0
Seedance 2.5 Flash is the quicker edition in Vellria's Seedance line. It shares the three modes, take lengths and generated sound of Seedance 2.5 and charges the same per second at the resolutions both offer; what it gives up is the top resolution step.
| Model | Price | Max resolution | Duration | Reference images |
|---|---|---|---|---|
| Seedance 2.5 Flash Uncensored This page | from 35 credits / s | 720p | 4–30s | Up to 1 |
| Seedance 2.5 Uncensored | from 35 credits / s | 1080p | 4–30s | Up to 1 |
| Seedance 2.0 Uncensored | from 30 credits / s | 1080p | 4–15s | Up to 1 |
Prices and limits come from the live model catalog.
Seedance 2.5 Flash or Seedance 2.5?
- Max resolution: 1080p (Seedance 2.5 Flash: 720p)
Move to Seedance 2.5 when the keeper has to be full HD. Prompts, start frames and reference pictures carry over unchanged, so a scene drafted on Flash can be rendered again there without rewriting it.
Seedance 2.5 Flash or Seedance 2.0?
- Price: from 30 credits / s (Seedance 2.5 Flash: from 35 credits / s)
- Max resolution: 1080p (Seedance 2.5 Flash: 720p)
- Duration: 4–15s (Seedance 2.5 Flash: 4–30s)
Seedance 2.0 is the earlier generation with shorter takes and a lower price per second. It suits scenes that fit a short take; it has no Character reference mode, so a scene assembled from several pictures stays on the newer models.
Seedance 2.5 Flash API: price and a first request
Seedance 2.5 Flash costs from 35 credits / s over the API, the same rate as in Studio. Model id vellria-seedance-2-5-flash on POST /v1/generate/video.
curl https://vellria.com/v1/generate/video \
-H "Authorization: Bearer vll_..." \
-H "Content-Type: application/json" \
-d '{
"model": "vellria-seedance-2-5-flash",
"prompt": "…",
"duration": 4,
"resolution": "480p",
"ratio": "adaptive"
}'Parameters, polling and uploads: API documentation · NSFW video API.
FAQ
Does Seedance 2.5 Flash allow NSFW?
Yes, for fictional adult content: nudity and explicit scenes between adults are allowed and the prompt goes out as you wrote it. Minors are refused automatically, and real people may not be animated or used as references without consent.
Is Seedance 2.5 Flash free?
Free to start: a new account created with Google gets 20 free credits, valid for 7 days (about 10 images on Z-Image Turbo Uncensored), and every signed-in account gets free daily uses of the NSFW prompt generator and the image to prompt tool. The gift can go toward Seedance 2.5 Flash at from 35 credits / s; beyond that every run is paid from prepaid credits at the rate this page shows before you start, and a run that fails at the provider is refunded automatically. Is there a free uncensored AI generator? covers the cheapest way to a first result.
What is the difference between Seedance 2.5 and Seedance 2.5 Flash?
Flash is the quicker edition of the same model, with the same three modes, take lengths and generated sound. It renders at 480p or 720p and costs the same per second as Seedance 2.5 at those steps; the full model adds 1080p.
How do reference images work in Seedance 2.5 Flash?
Choose Character reference as the reference mode, attach up to ten pictures and cite them in the prompt as @image1, @image2 and so on, in upload order. Each one is a person, object or place to feature rather than an opening frame, and extra pictures do not change the price.
Does Seedance 2.5 Flash generate sound?
Yes. Audio follows your description and is on by default; turn it off in the controls when you want a silent clip.
Is there a Seedance 2.5 Flash API?
Yes. The API section on this page holds a ready request with this model's id, and the API reference explains reference_mode, polling and downloads.






