Wan 3.0 Prime Reference to Video
Generate video from up to 10 reference images, 5 reference videos and 5 audio clips with Alibaba's Wan 3.0 Prime, the accelerated Wan 3.0. People, objects and styles from your references stay consistent through a continuous take with native audio. Reference images and audio are free; reference videos are billed as extra seconds.
12% less than the priciest provider (4 compared). Same model, same output.
Up to 10 reference images contributing people, objects, or styles. Cite in the prompt as Image 1, Image 2, ... At least one reference (image, video or audio) is required.
Up to 5 reference clips (mp4/mov, 1-15s each, 15s combined) for a subject's look and voice, motion, choreography or camera work. Cite as Video 1, ... Their length is billed as extra seconds, and reference + output must stay within 30s.
Up to 5 audio clips (wav/mp3, 1-15s each, 15s combined): a voice or song to lip-sync, dance or time the motion to. Cite as Audio 1, ... Free.
Resolution
Output resolution. 480p is the budget rung; 1080p is the premium rung.
Aspect Ratio
Aspect ratio of the generated video. 'adaptive' lets the model choose the best framing.
Duration
Length of the generated video in seconds (2-30). Billed per second. Wan 3.0 Prime generates up to 30 seconds in a single take.
Audio
Generate synchronized audio (dialogue, effects, music) in the same pass. Same price either way.
12% less than the priciest provider (4 compared). Same model, same output.
Your result renders here.
Use Wan 3.0 Prime Reference to Video via the API
Generate with Wan 3.0 Prime Reference to Video from your own code. Async by design: submit a generation, get a run id back instantly, poll until it completes. Billed in credits at the best available provider price — from $0.0612 per second.
1 — Start a generation
curl -X POST https://app.scenetra.com/api/v1/generate/wan-3.0-prime-ref2v \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt":"The person from the first image walks through the street from the second, handheld follow shot, crowd murmur","reference_image_urls":["https://example.com/person.jpg","https://example.com/street.jpg"],"duration":"5"}'Response (202)
{
"runId": "run_abc123",
"status": "PENDING",
"model": "wan-3.0-prime-ref2v",
"estimatedCredits": 630,
"statusUrl": "/api/v1/runs/run_abc123"
}2 — Poll until COMPLETED (every 2-5 seconds)
curl https://app.scenetra.com/api/v1/runs/run_abc123 \
-H "Authorization: Bearer YOUR_API_KEY"Response
{
"runId": "run_abc123",
"status": "COMPLETED",
"progress": { "completed": 1, "total": 1 },
"outputs": [
{ "type": "video", "url": "https://images.scenetra.com/..." }
],
"error": null,
"estimatedCredits": 630
}Request fields
| Field | Type | Description | Default |
|---|---|---|---|
| prompt* | string | Describe the video: subject, motion, camera work, and audio direction. Native synchronized audio is generated in the same pass. | — |
| reference_image_urls | string[] (https URLs) | Up to 10 reference images contributing people, objects, or styles. Cite in the prompt as Image 1, Image 2, ... At least one reference (image, video or audio) is required. | — |
| reference_video_urls | string[] (https URLs) | Up to 5 reference clips (mp4/mov, 1-15s each, 15s combined) for a subject's look and voice, motion, choreography or camera work. Cite as Video 1, ... Their length is billed as extra seconds, and reference + output must stay within 30s. | — |
| reference_audio_urls | string[] (https URLs) | Up to 5 audio clips (wav/mp3, 1-15s each, 15s combined): a voice or song to lip-sync, dance or time the motion to. Cite as Audio 1, ... Free. | — |
| resolution | string | Output resolution. 480p is the budget rung; 1080p is the premium rung.480p · 720p · 1080p | 720p |
| aspect_ratio | string | Aspect ratio of the generated video. 'adaptive' lets the model choose the best framing.adaptive · 16:9 · 4:3 · 1:1 · 3:4 · 9:16 | adaptive |
| duration | string | Length of the generated video in seconds (2-30). Billed per second. Wan 3.0 Prime generates up to 30 seconds in a single take.2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21 · 22 · 23 · 24 · 25 · 26 · 27 · 28 · 29 · 30 | 5 |
| audio | string | Generate synchronized audio (dialogue, effects, music) in the same pass. Same price either way.true · false | true |
| callbackUrl | string (https URL) | Optional. When the run finishes (completed or failed), we POST the final run status to this URL — same shape as the poll response, plus model, params, and chargedCredits. Best-effort; polling remains the source of truth. | |
Unknown fields are rejected with a helpful 400. Media fields take https URLs. There is no provider parameter — every run is routed to the cheapest capable provider automatically.
Good to know
Billing
Runs bill your account credits at provider cost with 0% markup. The submit response includes the credit estimate; failed generations are refunded automatically. Add ?dryRun=1 to get the estimate without running anything.
Errors
Errors are always { "error": { "code", "message" } } with stable codes: unauthorized, model_not_available, invalid_parameter, insufficient_credits, rate_limited. A failed run is not an HTTP error — poll responses report status: "FAILED" with the reason.
What makes Wan 3.0 Prime Reference to Video different
Wan 3.0, up to seven times faster
Prime runs the full Wan 3.0 toolset (text, first-frame and reference generation, native audio, 30-second takes) on an accelerated pipeline. Choose it when turnaround matters: iterating on a long take, previewing variations for a client, or keeping a production queue moving.
Single takes up to 30 seconds
Any length from 2 to 30 seconds in one continuous pass, with no stitching. That is long enough for a complete ad spot, a full scene or a product walkthrough, generated as a single shot.
Native audio at no extra cost
Dialogue, sound effects and music are generated in sync with the picture and steered by your prompt. Audio is on by default and the price is the same with it on or off.
Billed at cost in a visual workflow
Wan 3.0 Prime runs as a node in Scenetra's node-based editor, next to image and video models on one canvas. Every render goes to the cheapest provider that supports your settings and is billed at provider cost with 0% markup.
Parameters
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt* | text | Describe the video: subject, motion, camera work, and audio direction. Native synchronized audio is generated in the same pass. | — |
| Resolution | select | Output resolution. 480p is the budget rung; 1080p is the premium rung.480p · 720p · 1080p | 720p |
| Aspect Ratio | select | Aspect ratio of the generated video. Auto lets the model choose the best framing for your prompt.Auto · 16:9 · 4:3 · 1:1 · 3:4 · 9:16 | Auto |
| Duration | select | Length of the generated video in seconds. Wan 3.0 Prime generates any length from 2 to 30 seconds in a single take.2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21 · 22 · 23 · 24 · 25 · 26 · 27 · 28 · 29 · 30 | 5 |
| Audio | boolean | Generate synchronized audio (dialogue, SFX, music) in the same pass. Same price with audio on or off. | true |
Pricing
From $0.0612 per second
Pay per generation — only pay for what you use.
480p
$0.0612/s
per second
720p
$0.126/s
per second
1080p
$0.252/s
per second
What a video costs
| Duration | 480p | 720p | 1080p |
|---|---|---|---|
| 5 seconds | $0.31 | $0.63 | $1.26 |
| 10 seconds | $0.61 | $1.26 | $2.52 |
| 30 seconds | $1.84 | $3.78 | $7.56 |
Use cases
Marketing & ads
- Complete 30-second spots in one take
- Product teasers with synced sound
- Vertical 9:16 social ads
- Multi-length campaign variants
Social content
- Sound-on shorts and reels
- Long-form single-shot storytelling
- Ambient mood loops
- Square and portrait cuts from one prompt
Film & previz
- Full-scene previz without stitching
- Dialogue and SFX scratch tracks
- Establishing shots with ambience
- Cheap 480p blocking passes
Education & explainers
- Narrated explainer segments
- Step-by-step process clips
- Concept visualizations with sound
- Course intro sequences
Frequently asked questions
What is Wan 3.0 Prime reference-to-video?+
It's the reference mode of Alibaba's Wan 3.0 Prime: instead of one starting frame, the model uses your reference images, clips and audio to keep subjects, motion or voices consistent, and generates faster than standard Wan 3.0. Through the API it's wan-3.0-prime-ref2v.
How long can reference videos be?+
Up to 5 clips, 1 to 15 seconds each and 15 seconds combined. The reference length plus the output length must stay within 30 seconds, and reference seconds are billed at the same per-second rate as the output.
Related models
Wan 3.0
Alibaba's Wan 3.0 generates up to 30 seconds of video in a single take, with native synchronized audio — dialogue, sound effects, and music — produced in the same pass as the picture. Pick any duration from 2 to 30 seconds, output at 480p, 720p, or 1080p, and choose from five aspect ratios or let the model frame the shot automatically.
View model →Seedance 2.5
ByteDance's Seedance 2.5 generates cinematic videos up to 30 seconds from text prompts with synchronized audio — voice, sound effects, and background music — built in by default. The newest Seedance generation, with mp4 or high-fidelity mov output for multi-round editing pipelines.
View model →Kling 3.0 Pro
Kuaishou's flagship video generation model delivering stunning visual fidelity. Kling 3.0 Pro supports text-to-video, image-to-video with start/end frames, element references for character consistency, and native audio generation.
View model →Start creating with Wan 3.0 Prime Reference to Video
Use Wan 3.0 Prime Reference to Video alongside 50+ other AI models in Scenetra's visual workflow editor. No setup required.
Get Started