Seedance 2.0 Uncensored
Seedance 2.0 Uncensored is ByteDance's video model, hosted on Vellria with nothing to install: text to video and image to video, 480p to 1080p, clips of 4–15s with sound. Run it in Studio or over the REST API, from 30 credits / s (100 credits = $1.00).
vellria-seedance-2-0Seedance 2.0 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 yetCover art
from 30 credits / s
Your request will cost 150 credits.
About Seedance 2.0
Also known as: Seedance 2 · ByteDance Seedance 2.0
Seedance 2.0 is the ByteDance video model most people know the Seedance name from, built by the Seed team that also makes the Seedream image models. It is the generation before Seedance 2.5, and Vellria runs both. Plenty of searches still ask for Seedance 2.0 by name, so this page runs that exact model rather than sending you to its successor: the same family habits and prompt style, in the earlier release.
Clips come back with sound generated from your description, and audio is on by default; switch it off in the controls when you plan to score the clip yourself. Takes are shorter than on 2.5, topping out at fifteen seconds, and each second costs less, which makes 2.0 a sensible place to work out staging, pacing and spoken lines before spending on a longer take. Describe the action as a single sequence in the order it happens; a take reads better as one movement than as a list of unrelated beats.
Text-to-video offers a wide choice of proportions, from cinemascope and landscape through square to vertical portrait, plus an adaptive setting that lets the model decide. Image-to-video takes one picture and turns it into the first frame of the clip; the output follows that picture's proportions rather than a ratio from the menu, so prepare the crop in advance. Stills from Seedream 5.0 Pro are a natural source, since both models come from ByteDance's Seed team.
Resolution comes in three steps, and both resolution and length raise the price, so draft at the lowest step with a short duration, then re-run the settled prompt higher up. A re-run is a new take rather than an upscale, so expect small differences between drafts and the keeper. When a scene outgrows fifteen seconds, move the same prompt to the newer model; for a cheap scratchpad that has its own finishing tier, MiniMax H3 Video is the alternative to compare.
Seedance 2.0 in action
Seedance 2.0 outputs with the exact prompts behind them.
Image to video
She lifts the whiskey glass, takes a slow sip and glances toward the camera as neon light flickers through the window, subtle slow push-in
Image to video
She tucks her hair behind her ear and smiles softly at the camera while staying seated, warm lamplight flickering, slow gentle push-in on her face
Text to video
A man and a woman in their thirties dance a slow tango in an empty candlelit ballroom, her red dress swirling, smooth cinematic orbit around them
When to pick Seedance 2.0
Pick it when
- Your scene fits within a fifteen-second take and you want Seedance motion with sound at a lower cost per second than 2.5.
- You are working out staging, pacing and dialogue on short takes before committing a final version to a longer or pricier model.
- You need a cinemascope or vertical portrait clip from text, or want the adaptive setting to choose the proportions for you.
Pick another model when
- The shot must run longer than fifteen seconds in one take; Seedance 2.5 is the newer version with longer takes.
- You want to test a camera move cheaply before a separate finishing tier; MiniMax H3 Video pairs with H3 Max for that.
- Your clip has to continue an image you made without reinterpretation; MiniMax H3 Max stayed close to the supplied frame in our runs.
Prompt or reference
For staging, starting from text works well here: describe the action, let the adaptive setting pick the proportions or name a frame shape, and keep the take short. Upload a still when a character or a room must carry into motion; the picture opens the clip and its own proportions decide the frame.
How to use Seedance 2.0 on Vellria
- Prompt: up to 10,000 characters in the panel above.
- Cost: from 30 credits / s — see pricing.
Example prompts
An adult woman in a satin robe pours two glasses of wine in a dim kitchen, glances toward the doorway and smiles, low hum of a fridge, soft jazz from another room.
From the uploaded still: the adult man on the hotel balcony sets down his coffee, unbuttons his shirt in the morning breeze and leans on the railing, distant traffic and seagulls.
Cinemascope frame: two adults lie in tall grass at sunset, she rolls onto her side and whispers something that makes him laugh, crickets and wind.
Vertical portrait frame: an adult dancer in a mesh bodysuit rehearses a slow routine in a mirrored studio, bare feet on the wooden floor, a metronome ticking.
Short take: an adult couple shelter from a downpour in a doorway, kiss once and run out into the rain laughing, thunder rumbling, splashing footsteps.
Tips
- Keep each take to one continuous action; with a fifteen-second ceiling, a single movement plays better than several crammed beats.
- Name the sounds you want, such as rain, breathing or a record player, or turn audio off if you will add your own track.
- Draft at the lowest resolution and a short length, then re-run the same wording at a higher step for the version you keep.
- For image-to-video, crop the picture to the proportions the clip should have; the ratio menu applies only to text-to-video.
Specifications
| Resolution: | 480p720p1080p |
|---|---|
| Duration: | 4–15s |
| 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 | 30 credits / s |
| 720p | 60 credits / s |
| 1080p | 140 credits / s |
Seedance 2.0 vs Seedance 2.5
Seedance 2.0 is the earlier and lower-priced of the two Seedance versions on Vellria. It keeps the same resolution steps, frame shapes and generated sound as 2.5; the difference is how long a single take can run.
| Model | Price | Max resolution | Duration | Reference images |
|---|---|---|---|---|
| Seedance 2.0 Uncensored This page | from 30 credits / s | 1080p | 4–15s | Up to 1 |
| Seedance 2.5 Uncensored | from 35 credits / s | 1080p | 4–30s | Up to 1 |
Prices and limits come from the live model catalog.
Seedance 2.0 or Seedance 2.5?
- Price: from 35 credits / s (Seedance 2.0: from 30 credits / s)
- Duration: 4–30s (Seedance 2.0: 4–15s)
Switch to Seedance 2.5 when a shot needs to run past what 2.0 allows in one take. The prompt and start frame carry over unchanged; the extra length costs more per second.
Seedance 2.0 API: price and a first request
Seedance 2.0 costs from 30 credits / s over the API, the same rate as in Studio. Model id vellria-seedance-2-0 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-2-0",
"prompt": "…",
"duration": 4,
"resolution": "480p",
"ratio": "adaptive"
}'Parameters, polling and uploads: API documentation · NSFW video API.
FAQ
Does Seedance 2.0 allow NSFW?
Yes, for fictional adult content: nudity and explicit scenes between adults are allowed and the prompt is not rewritten. Minors are refused automatically, and real people may not be animated without consent.
Is Seedance 2.0 free?
No, there is no free plan, but a new account created with Google gets 20 credits as a gift, valid for 7 days, which you can put toward Seedance 2.0 at from 30 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.
Is this Seedance 2.0 or Seedance 2.5?
This page runs Seedance 2.0, the earlier ByteDance release. Seedance 2.5 is the newer model with longer takes and has its own page; prompts move between the two without changes.
How long can a Seedance 2.0 clip be?
Up to fifteen seconds per take. The specifications table on this page lists the exact range and the price per second at each resolution, read from the live catalog.
Does Seedance 2.0 generate sound?
Yes. Audio is generated from your description and is on by default; you can turn it off when you want a silent clip.
Can I use Seedance 2.0 through an API?
Yes. Copy the request from the API section on this page, which already names Seedance 2.0, and follow the API reference to poll the job and download the clip.
Related models
Make the start frame for your clip with one of our image models.
Seedream 5 Lite Uncensored
Large 3K–4K images, 21:9 ultrawide included; edit or build a series from up to 10 references.
Image
8 credits
Qwen Image 2.1 Uncensored
Uncensored 1K images in seven shapes; edit or merge up to 10 references, each extra one adding a little.
Image
from 7 credits
Wan 2.7 Uncensored
Crisp 2K images that follow your wording; edit or blend up to 9 photos in their own shape.
Image
8 credits