Seedance 2 unfiltered is a search this key cannot answer with Seedance itself: Seedance 2 ships through its developer with a filter, and the Seedance rows on this key carry no switch to turn it off. What this key does have is a set of uncensored video builds, on the same endpoint and the same balance. This post puts both sets in one place, prices every row per second from the live catalog, explains what R2V means, and names the rows that take reference images.
The per-clip price breakdown and the request loop are in the uncensored video generator API post; the uncensored image and video models are on the media page. Venice's Seedance page, seen in the 27 October 2026 results, is on venice.ai.
What "seedance 2 unfiltered" turns up, and what it means here
The Google results for the phrase on 27 October 2026 were a Reddit thread asking for the best unfiltered Seedance 2 service, then vendor pages from ZenCreator, VideoAny, Wiro, Apipod and Kenerate with "uncensored Seedance" in their titles, Venice's Seedance 2.0 page, and a Sogni post. Four ads on top.
Here is the plain version of what this key does. Seedance 2, 2 Fast, 2 Mini and 2.5 are all on the catalog, by API, priced per second. Their rows carry no enable_safety_checker setting, so there is no switch to send; the clip goes through the developer's filter. The uncensored video on this key is a different set of models, listed in the second table, and they take the same request shape. This post prices both sets from the live catalog so you can see what each route costs before you pick one.
Seedance 2 and 2.5 on this key: the filter stays on
The Seedance rows from GET /v1/catalog/media on 2026-11-09. The price is per second of finished video and depends on resolution. The last column quotes the row: where a model has a safety-checker setting, the column shows its default; where it has none, it says so.
| Model | Settings | Per second | 5 s clip | 10 s clip | Filter, per the row |
|---|---|---|---|---|---|
| Seedance 2 | resolution 480p | $0.125 | $0.63 | $1.25 | no enable_safety_checker setting on the row |
| resolution 720p | $0.25 | $1.25 | $2.50 | ||
| resolution 1080p | $0.5625 | $2.81 | $5.63 | ||
| resolution 4k | $1.25 | $6.25 | $12.50 | ||
| Seedance 2 Fast | resolution 480p | $0.0875 | $0.44 | $0.88 | no enable_safety_checker setting on the row |
| resolution 720p | $0.175 | $0.88 | $1.75 | ||
| Seedance 2 Mini | resolution 480p | $0.0625 | $0.31 | $0.63 | no enable_safety_checker setting on the row |
| resolution 720p | $0.15 | $0.75 | $1.50 | ||
| Seedance 2.5 | resolution 480p | $0.56 | $2.80 | $5.60 | no enable_safety_checker setting on the row |
| resolution 720p | $1.26 | $6.30 | $12.60 | ||
| resolution 1080p | $2.28 | $11.40 | $22.80 |
Seedance 2 takes image_urls, reference_image_urls, reference_audio_urls and a generate_audio switch, with resolutions 480p, 720p, 1080p, 4k. A prompt the developer's filter declines comes back as a failed generation with the charge reversed, the same refund rule every media model on this key follows.
The honest line for a searcher who typed "uncensored seedance": a Seedance clip here is filtered, and this key has no setting that removes it. What this key sells instead is a separate set of uncensored video builds, each under its own name. That is the next table.
The uncensored video builds, priced per second
Each row is a video model served as an uncensored build. wan-3.0-video-uncensored is text to video; wan-3.0-i2v-uncensored and ltx-2.3-i2v-uncensored take an image as the first frame, so a still from an uncensored image model becomes a clip. The filter column again quotes the row: the checker's default where the row has the setting, and where an uncensored build has none, that it has none.
| Model | Settings | Per second | 5 s clip | 10 s clip | Filter, per the row |
|---|---|---|---|---|---|
| Wan 3.0 Video Uncensored | resolution 480p | $0.20 | $1.00 | $2.00 | enable_safety_checker defaults to false |
| resolution 720p | $0.40 | $2.00 | $4.00 | ||
| resolution 1080p | $0.80 | $4.00 | $8.00 | ||
| Wan 3.0 Image-to-Video Uncensored | resolution 480p | $0.20 | $1.00 | $2.00 | uncensored build with no checker setting |
| resolution 720p | $0.40 | $2.00 | $4.00 | ||
| resolution 1080p | $0.80 | $4.00 | $8.00 | ||
| LTX-2.3 Image-to-Video Uncensored | resolution 480p | $0.08 | $0.40 | $0.80 | uncensored build with no checker setting |
| resolution 720p | $0.16 | $0.80 | $1.60 | ||
| resolution 1080p | $0.24 | $1.20 | $2.40 |
The per-second breakdown for 5, 10 and 15 second clips, the submit-poll-download loop and what 100 clips cost are in the uncensored video generator API post. One request looks like this:
curl https://api.uncensoredllm.ai/v1/generations \ -H "Authorization: Bearer sk-ullm-live-…" \ -H "Content-Type: application/json" \ -d '{ "model": "wan-3.0-video-uncensored", "input": { "prompt": "a dancer on a rooftop at dusk, slow dolly in, city lights behind", "resolution": "480p", "duration": 5 } }'The 202 carries the charge, seconds times the rate, before a frame renders. To check the billed seconds against the file afterwards, ffprobe prints the duration:
ffprobe -v error -show_entries format=duration -of default=nw=1:nk=1 clip.mp4Seedance 2.0 R2V: what reference-to-video means
R2V is reference-to-video. Instead of one first frame, you hand the model reference images of a character, an object or a setting, and the prompt describes a new shot; the model keeps the references consistent through the clip. Image-to-video (I2V) animates a single picture from its first frame. Text-to-video (T2V) has no picture at all.
On this key the parameter that carries references is reference_image_urls, and image_urls is the first frame. What each row on this page takes, on 2026-11-09:
| Model | Image and audio inputs | From, per second |
|---|---|---|
| Seedance 2 | image_urls, reference_image_urls, reference_audio_urls | $0.125 |
| Seedance 2 Fast | image_urls, reference_image_urls, reference_audio_urls | $0.0875 |
| Seedance 2 Mini | image_urls, reference_image_urls, reference_audio_urls | $0.0625 |
| Seedance 2.5 | image_urls, reference_image_urls, reference_audio_urls | $0.56 |
| Wan 3.0 Video Uncensored | text prompt only | $0.20 |
| Wan 3.0 Image-to-Video Uncensored | image_urls (required) | $0.20 |
| LTX-2.3 Image-to-Video Uncensored | image_urls (required) | $0.08 |
{ "model": "seedance-2", "input": { "prompt": "the same character walks through a rain-lit market", "reference_image_urls": ["https://example.com/character.png"], "resolution": "720p", "duration": 5 }}The uncensored image-to-video rows take image_urls only: a first frame, not a reference set. So the uncensored route to a consistent character is an image model first, then I2V. Make the still on an uncensored image model from the image and video page, then pass its URL as image_urls to wan-3.0-i2v-uncensored or ltx-2.3-i2v-uncensored.
Every uncensored media model on this key
The video rows above and the uncensored image models they pair with, all on POST /v1/generations, priced per image or per second. The image side's price breakdown is in the uncensored image generator API post.
| Per | Endpoints | |||
|---|---|---|---|---|
| LTX-2.3 Image-to-Video Uncensored | Video | $0.08 | second | |
| Wan 3.0 Image-to-Video Uncensored | Video | $0.20 | second | |
| Wan 3.0 Video Uncensored | Video | $0.20 | second | |
| Anime Image Uncensored | Images | $0.03 | image | |
| Chroma Uncensored | Images | $0.08 | image | |
| CyberRealistic XL Uncensored | Images | $0.0136 | image | |
| HunyuanImage 3.0 Instruct Uncensored | Images | $0.1125 | image | |
| Qwen Edit Uncensored | Images | $0.08 | image | |
| Qwen Image 2.1 Uncensored | Images | $0.0375 | image | |
| Qwen Image 3 Uncensored | Images | $0.08 | image | |
| Seedream 5.0 Lite Edit Uncensored | Images | $0.0313 | image | |
| Vision Image Uncensored | Images | $0.04 | image | |
| Vision Pro Image Uncensored | Images | $0.09 | image | |
| WAI Illustrious SDXL Uncensored | Images | $0.0136 | image |
Questions
Not on this key. Seedance 2 is served through its developer with a content filter and no public weights, and the rows on this key carry no safety-checker setting to send. The uncensored video on this key is a different set of models: wan-3.0-video-uncensored and the two image-to-video rows, priced per second in the table. See the image and video page.
No. There are no public Seedance 2 weights to download, so any "download" result is a web app or a different model. The uncensored video builds on this key run by API with no GPU.
The Seedance rows go through the developer’s filter, with the charge reversed on a declined clip. The uncensored video rows are separate builds, and the table shows each row’s checker setting. On every uncensored image and video model, a prompt that describes a minor is refused before it is priced, and the image-to-video rows animate only images generated here. The acceptable use policy has the full list. See the acceptable use policy.
Reference-to-video: reference images (and on Seedance 2, reference audio) sent next to the prompt, which the model keeps consistent through the clip. On this key the field is reference_image_urls; the table above shows which rows on this page take it.
Not here. Credit is prepaid from $10, every clip is priced per second before it renders, and a clip that fails is refunded in full. See what 100 clips cost.
Venice lists a Seedance 2.0 page, which appeared in the 27 October 2026 results. This site does not reproduce its claims; read them at venice.ai. Here Seedance 2 runs with the developer's filter, and the uncensored clips come from the uncensored rows above.
Every model above is on the uncensored model list with its live price per million tokens. Per-token pricing sets the whole catalog side by side, and the family directory groups the builds by base model.