APIPod logoAPIPod
Alibaba

Wan 3.0 Video Prime Spicy

3 models available
Provider: Alibaba
Updated: Recently

Wan 3.0 Video Prime Spicy is the faster tier of Wan 3.0 Spicy: the same contract - 2-30 second clips with native audio at 480P, 720P, or 1080P, first/last-frame control, and rich reference materials - on a quicker relaxed-moderation route, at a higher per-second rate.

Available Models

18+Wan 3.0 Prime Spicy API · T2V · I2V · R2V

Same stock.Rush processing.

Wan 3.0 Prime Spicy is the fast tier of Wan 3.0 Spicy: Alibaba’s Wan 3.0 video model with relaxed moderation for adult (18+) work, on a quicker upstream route. Same request, same 480P–1080P output, same 2–30 second clips with native audio — you pay a higher rate per second to get the take back sooner.

R2V · 16:9 · RUSH

Lab order

wan3.0-prime-spicy-*

Stock
w3.0-video · Spicy
Process
T2V · I2V · R2V
Resolution
480P · 720P · 1080P
Length
2–30 s, or -1
Sound
Native, in sync
Prompt
prompt_extend

Turnaround

Fast tier — ahead of standard processing

Rush surcharge

Same roll. Pay for the clock.

Prime Spicy bills exactly like Spicy — rate × (reference video seconds + output seconds) — so the only thing you pay extra for is speed. Pick a resolution and a length to see what rush processing adds to one take.

+67%

more per second than Wan 3.0 Spicy at the same resolution. That is the whole rush fee.

Resolution

Standard processing

wan3.0-spicy-*

$2.40

Rush processing

wan3.0-prime-spicy-*

$4.00

Rush adds $1.60 to this take

Rates are APIPod’s live list prices for the Wan 3.0 Spicy and Prime Spicy models. Turnaround depends on upstream load and is not guaranteed. Failed tasks are refunded automatically.

When to rush

Rush the takes someone is waiting for.

Speed is worth paying for when a person is watching the progress bar. Anything that can wait for morning should run at the standard rate.

Rush

Someone is waiting

In-app generation, chat companions and preview buttons, where every second of waiting is a user deciding whether to stay.

wan3.0-prime-spicy-*

Rush

The cut is due today

Client reviews, campaign drops and fix-ups after notes, when the next round of feedback is booked before a standard queue would clear.

wan3.0-prime-spicy-*

Standard

It can wait for morning

Batch jobs, back-catalogue renders and prompt sweeps. Nobody is watching, so run them on Wan 3.0 Spicy and keep the difference.

wan3.0-spicy-*

Edit decision list

Three model IDs, one endpoint.

Pick the model by what you bring: a prompt, a pair of keyframes or a box of references. All three take resolution 480P, 720P or 1080P, plus the same duration, aspect_ratio, generate_audio and prompt_extend fields as Wan 3.0 Spicy.

Model IDInputLimitsUse it for
wan3.0-prime-spicy-t2vPrompt onlyprompt up to 20,000 charactersScenes written from scratch, when the answer is needed now.
wan3.0-prime-spicy-i2vPrompt + first frame, optional last frameimage_urls: 1–2 · first_frame_image / last_frame_imageBringing a still to life inside an interactive product.
wan3.0-prime-spicy-r2vPrompt + references named in the prompt10 images · 5 videos · 5 audio · 1 file · 1 linkRecurring characters and voices on a tight turnaround.

Spec sheet

Wan 3.0 and three Spicy tiers, side by side.

Prime Spicy is Spicy on rush processing: the same request, resolutions and moderation, with a faster upstream tier and a higher per-second rate.

SpecWan 3.0Wan 3.0 SpicyWan 3.0 Prime SpicyWan 3.0 Pro Spicy
PositioningStandard routeStandard speed, lowest rateFast tier — quicker turnaround, higher rateResolution tier — 2K / 4K
ModerationStandard upstream reviewRelaxed, for adult (18+) workRelaxed, for adult (18+) workRelaxed, for adult (18+) work
Resolution480P · 720P · 1080P480P · 720P · 1080P480P · 720P · 1080P1080P · 2K · 4K
Clip length2–30 s, or -1 for smart length2–30 s, or -1 for smart length2–30 s, or -1 for smart length2–30 s, or -1 for smart length
AudioNative, generated with the pictureNative, generated with the pictureNative, generated with the pictureNative, generated with the picture
References (R2V)10 images · 5 videos · 5 audio · file · link10 images · 5 videos · 5 audio · file · link10 images · 5 videos · 5 audio · file · link10 images · 5 videos · 5 audio · file · link
prompt_extendNot availableSupported — set false for literal promptsSupported — set false for literal promptsSupported — set false for literal prompts
Keyframes (I2V)image_urls: first + optional lastimage_urls: first + optional lastimage_urls: first + optional lastimage_urls only: first + optional last
AvailabilityAll APIPod regionsNot in mainland China, Hong Kong or MacauNot in mainland China, Hong Kong or MacauNot in mainland China, Hong Kong or Macau

Call sheet

Swap one string. Get it back sooner.

The same asynchronous /v1/videos/generations endpoint as every video model on APIPod. This example writes an 8-second vertical take at 1080P; moving an existing Spicy integration over is a model-string change.

  • Every Spicy field works unchanged: resolution, aspect_ratio, duration, generate_audio, seed and prompt_extend.
  • Keyframes go in image_urls (first frame, optional last frame) or first_frame_image / last_frame_image.
  • duration: -1 lets the model pick the length. It is reserved at 30 s and settled on the actual output.
  • Route by urgency: send user-facing requests to wan3.0-prime-spicy-* and batch work to wan3.0-spicy-*.
Read the Prime Spicy docs
curl https://api.apipod.ai/v1/videos/generations \
  -H "Authorization: Bearer $APIPOD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan3.0-prime-spicy-t2v",
    "prompt": "Rain-slick rooftop at 2 a.m.; she vaults the ledge onto a fire escape as police lights sweep the alley below. Handheld tracking shot, 24mm, sodium haze, sirens and heavy breath.",
    "resolution": "1080P",
    "aspect_ratio": "9:16",
    "duration": 8,
    "generate_audio": true
  }'

# → { "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

Prompts for takes that cannot wait.

Short, kinetic scenes for adult-oriented genre work — the kind you iterate on while someone is in the room. Copy one and recast it.

  • Rooftop chaseT2V · 1080P · 9:16 · 8 s

    Rain-slick rooftop at 2 a.m.; she vaults the ledge onto a fire escape as police lights sweep the alley below. Handheld tracking shot, 24mm, sodium haze, sirens and heavy breath.

  • Last callI2V · 720P · 10 s

    The bartender in the first frame slides a glass down the bar, leans in close and says: “We’re closed — unless you’re staying.” Slow push-in, red neon through smoke, low bass from the jukebox.

  • HeistR2V · 1080P · 12 s

    Image 1 cracks the vault door in Video 1’s corridor while alarms strobe red; she turns to camera with the stones in her fist. Whip pan, 35mm, sparks and a pounding score.

  • StormT2V · 480P · -1

    A lighthouse keeper fights a door against a hurricane; the lamp swings overhead and a wave explodes over the gallery. Wide to close in one move, salt spray on the lens, roaring wind.

House rules · 18+

Adults only
Spicy models are for users 18 and over, or the age of majority where you live.
Never minors
Any sexualised depiction of minors — real, fictional or implied — is prohibited.
Real people need consent
No intimate or defamatory content of identifiable real people without their explicit consent.
Region limits
Not available in mainland China, Hong Kong or Macau; API requests from those regions are rejected.

Questions

Before you rush it.

What is Wan 3.0 Prime Spicy?

Wan 3.0 Prime Spicy is the fast tier of Wan 3.0 Spicy on APIPod: the same relaxed-moderation Wan 3.0 video model for adult (18+) work, served from a quicker upstream route at a higher per-second rate. It covers text-to-video, keyframe-to-video and reference-to-video in 2–30 second clips with native audio at 480P, 720P or 1080P, through wan3.0-prime-spicy-t2v, wan3.0-prime-spicy-i2v and wan3.0-prime-spicy-r2v.

How is it different from Wan 3.0 Spicy?

Turnaround and price only. The request contract is identical field for field — modes, resolutions, reference limits, clip length, audio, prompt_extend and moderation all match. Prime Spicy runs on the upstream’s fast tier and costs more per second.

How much faster is it?

Prime Spicy is the upstream’s faster tier of the same model, so takes usually come back sooner than on Wan 3.0 Spicy. Actual turnaround depends on clip length, resolution, references and upstream load, so APIPod does not promise a fixed number of seconds — measure it on your own prompts before routing production traffic.

How is Prime Spicy billed?

Per second, the same way as Spicy: the resolution’s rate × (reference video seconds + output seconds). Only the rate is higher. Audio on or off costs the same, and failed tasks are refunded automatically.

How do I move a Spicy integration to Prime Spicy?

Change the model string — wan3.0-spicy-t2v becomes wan3.0-prime-spicy-t2v, and the same for i2v, r2v and the all-in-one wan3.0-video-prime-spicy. No other field changes, so you can route urgent requests to Prime Spicy and keep batch work on Spicy with one switch.

How long can a Prime Spicy clip be?

Any whole number of seconds from 2 to 30 in a single pass. Pass duration -1 to let the model choose; it is reserved at 30 seconds and settled at the actual length.

Where can I use Wan 3.0 Prime Spicy?

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.

Ready when you are.

Create a key, call wan3.0-prime-spicy-t2v and get your first rush take back. Same API and the same key as every other model on APIPod.

Get an API key