Wan 3.0 Video Spicy keeps the Wan 3.0 contract - 2-30 second clips with native audio, first/last-frame control, and rich reference materials - on a relaxed-moderation route for adult-oriented creative work. Pricing runs slightly above the standard models.
Wan 3.0 Spicy T2V on APIPod keeps the Wan 3.0 text-to-video contract - 2 to 30 second clips, up to 1080P, native audio - and routes to a relaxed-moderation upstream built for mature-theme creative work. Pricing is slightly above the standard model.
Wan 3.0 Spicy I2V on APIPod animates a first frame (optionally with a last frame) into a 2 to 30 second clip with native audio, up to 1080P, on a relaxed-moderation upstream built for mature-theme creative work. Pricing is slightly above the standard model.
Wan 3.0 Spicy R2V on APIPod generates 2 to 30 second clips guided by reference images, videos, audios, and a file or link, with native audio up to 1080P, on a relaxed-moderation upstream built for mature-theme creative work. Pricing is slightly above the standard model.
18+Wan 3.0 Spicy API · T2V · I2V · R2V
Wan 3.0 Spicy is Alibaba’s Wan 3.0 video model with relaxed moderation instead of the standard content filter: 2–30 second clips with native audio at 480P, 720P or 1080P, for adult (18+) creative work. Iterate cheaply at 480P, then print the keeper at 1080P on the same API.
Dailies math
Most takes are tests: timing, blocking, whether the prompt lands at all. Spend those seconds at the low rate and pay the full rate only for the take you keep. Pick a budget and a clip length — every frame below is one complete generation.
×4
drafts at 480P cost the same as one take at 1080P.
480P
Prompt tests, blocking, timing
16 takes
$0.60 per take
720P
Previews and social feeds
8 takes
$1.20 per take
1080P
The take you deliver
4 takes
$2.40 per take
Rates are APIPod’s live list prices for the Wan 3.0 Spicy models. Requests with reference videos are billed on reference seconds plus output seconds. Failed tasks are refunded automatically.
Edit decision list
Pick the model by what you bring to set: a prompt, a pair of keyframes or a box of references. All three post to /v1/videos/generations and accept the same duration, resolution, aspect_ratio, generate_audio and prompt_extend fields.
| Model ID | Input | Limits | Use it for |
|---|---|---|---|
| wan3.0-spicy-t2v | Prompt only | prompt up to 20,000 characters | Writing a scene from scratch: camera, light, dialogue, pacing. |
| wan3.0-spicy-i2v | Prompt + first frame, optional last frame | image_urls: 1–2 | Locking a look, a character or a product shot before it moves. |
| wan3.0-spicy-r2v | Prompt + references named in the prompt | 10 images · 5 videos · 5 audio · 1 file · 1 link | Casting from references: faces, wardrobe, voices, motion. |
Spec sheet
One model family, three routes. What changes is moderation, resolution and where you can call them from; the request format stays the same.
| Spec | Wan 3.0 | Wan 3.0 Spicy | Wan 3.0 Pro Spicy |
|---|---|---|---|
| Moderation | Standard upstream review | Relaxed, for adult (18+) work | Relaxed, for adult (18+) work |
| Resolution | 480P · 720P · 1080P | 480P · 720P · 1080P | 1080P · 2K · 4K |
| Clip length | 2–30 s, or -1 for smart length | 2–30 s, or -1 for smart length | 2–30 s, or -1 for smart length |
| Audio | Native, generated with the picture | Native, generated with the picture | Native, generated with the picture |
| References (R2V) | 10 images · 5 videos · 5 audio · file · link | 10 images · 5 videos · 5 audio · file · link | 10 images · 5 videos · 5 audio · file · link |
| prompt_extend | Not available | Supported — set false for literal prompts | Supported — set false for literal prompts |
| Keyframes (I2V) | image_urls: first + optional last | image_urls: first + optional last | image_urls only: first + optional last |
| Availability | All APIPod regions | Not in mainland China, Hong Kong or Macau | Not in mainland China, Hong Kong or Macau |
Call sheet
The same asynchronous /v1/videos/generations endpoint as every video model on APIPod: submit, poll, download. This example animates a first frame at 720P and keeps the prompt literal.
curl https://api.apipod.ai/v1/videos/generations \
-H "Authorization: Bearer $APIPOD_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "wan3.0-spicy-i2v",
"prompt": "She lowers the vintage camera, looks straight into the lens and smiles; neon rain streaks the window behind her. Handheld 35mm, shallow focus, soft synth hum.",
"image_urls": ["https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20260408/sjuytr/wan-r2v-object-girl.jpg"],
"resolution": "720P",
"duration": 10,
"prompt_extend": false
}'
# → { "code": 200, "data": { "task_id": "…", "status": "pending" } }
curl https://api.apipod.ai/v1/videos/status/$TASK_ID \
-H "Authorization: Bearer $APIPOD_API_KEY"Shot list
Starting points for adult-oriented genre work, short enough to draft at 480P and strong enough to print at 1080P. Copy one, recast it, roll.
Vertical thrillerT2V · 720P · 9:16 · 8 s
Phone-camera POV climbing a dark stairwell; a hand pushes open a door marked 13 and the light behind it dies. Ragged breathing, footsteps on the floor above. Grainy high-ISO look, framed for a vertical feed.
Fashion shortI2V · 1080P · 10 s
From the first frame: the model rises from a velvet chaise in a black satin slip, crosses to the window and glances back over her shoulder. Warm tungsten practicals, 85mm, slow motion on the fabric, low lounge piano.
HorrorT2V · 480P draft · 6 s
Locked-off wide of a motel corridor at 3 a.m.; every door is shut except the last one, which opens an inch on its own. Fluorescent flicker, the hiss of a distant television, then silence.
Fight sceneR2V · 1080P · 15 s
Video 1 as the fighter, Image 1 as her opponent: a bare-knuckle fight in a flooded parking garage, every hit throwing up spray. Handheld, crisp fast-shutter look, sodium light, heavy impacts in the sound mix.
Questions
Wan 3.0 Spicy is the relaxed-moderation route to Alibaba’s Wan 3.0 video model on APIPod, for adult (18+) creative work. It generates 2–30 second clips with native audio at 480P, 720P or 1080P from text, keyframes or up to 20 reference files, through three models: wan3.0-spicy-t2v, wan3.0-spicy-i2v and wan3.0-spicy-r2v.
It runs with relaxed upstream moderation, so mature themes the standard route refuses — graphic horror, hard-hitting violence, sensual and intimate scenes — are accepted. It is not lawless: content that sexualises minors, non-consensual imagery of real people and illegal material are always refused.
Draft at 480P while you lock the prompt, camera and timing — it has the lowest per-second rate. Use 720P for previews and social feeds, and render the keeper at 1080P. If you need 2K or 4K, use Wan 3.0 Pro Spicy on the same API.
Any whole number of seconds from 2 to 30 in a single pass — double Wan 2.7’s 15-second ceiling. Pass duration -1 to let the model choose a length that fits the scene; it is reserved at 30 seconds and settled at the actual length.
Per second: the resolution’s rate × (reference video seconds + output seconds). Requests without reference videos are billed on output seconds only. Audio on or off costs the same, and failed tasks are refunded automatically.
They share modes, reference limits, clip length, audio and moderation. Spicy renders 480P to 1080P and is the cheaper tier for iteration and social video; Pro Spicy renders 1080P, 2K and 4K for large screens and for re-cutting one master into several formats.
Everywhere APIPod operates except mainland China, Hong Kong and Macau, where Spicy pages are unavailable and API requests are rejected. Use is limited to adults aged 18 and over.
Create a key, call wan3.0-spicy-t2v and send your first take. Same API and the same key as every other model on APIPod.
Get an API key