Vidu Q4 Reference-to-Video generates subject-consistent AI videos from text prompts, 1-15 reference images, and up to 3 reference audio clips, supporting character consistency, product videos, social content, brand assets, and reference-guided storytelling workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Idle
$0.095per run·~10 / $1
Ultra-realistic, cinematic Hollywood Western style, realistic live-action filming. Golden sunset at an American ranch, with wooden fences, sandy ground, and open wilderness in the distance. A sudden strong gust of wind sweeps across the ranch, lifting dust and sand into the air. The overall mood is rugged, cinematic, dramatic, and powerful. An adult Western cowboy stands in front of the ranch fence, keeping the same outfit and appearance: a brown cowboy hat, brown suede fringe jacket, white shirt, dark jeans, a wide leather belt, and a large metal belt buckle. He has realistic skin texture and a rugged, handsome presence. At first, he stands still for a brief moment. Then a powerful gust of wind suddenly rushes through the ranch, whipping up sand and dust around him. He immediately lowers his head slightly and presses one hand firmly onto the brim of his cowboy hat to keep it from blowing away. His jacket fringe, shirt hem, and clothes are pushed hard by the wind. He narrows his eyes against the blowing dust and begins walking forward into the wind with steady, determined steps. The gust continues as he moves ahead, creating a strong sense of resistance and motion. Dust swirls around his legs and across the ground, while his posture remains calm, strong, and controlled. He ends in a striking forward-moving pose, still holding his hat, walking against the wind like a Western film protagonist. Camera: Use a medium-shot cinematic handheld or slightly stabilized follow shot. The camera should subtly track backward as he walks forward, keeping him framed in a strong hero composition. No random cuts, no dramatic zooms. Focus on the movement of the wind, the blowing dust, and his confident forward motion. The scene should feel like a dramatic Western movie moment, not a fashion pose. Mood keywords: Western, windstorm, blowing dust, rugged, heroic, cinematic, tense, dramatic, strong, realistic.
Use the character reference image to create a high-energy cinematic fashion video. The same man stands in a dark industrial space, then slowly turns toward camera as a strong gust of wind moves his jacket and hair. He takes one confident step forward while sharp white lights flicker behind him, creating metallic reflections across the distressed leather jacket and sunglasses. Add subtle handheld camera movement, a quick low-angle push-in, dramatic shadows, smoky atmosphere, fast light streaks, and a powerful streetwear campaign mood. Keep his face, hairstyle, sunglasses, outfit, body proportions, and overall character design consistent with the reference image. Cool, edgy, futuristic, premium fashion-film aesthetic.
/filters:quality(82)/media/images/fd33d25b-4c45-4db2-8d60-ed2715dc0982.webp)
gemini-omni-1.2-flash/reference-to-video
Reference to Video
/filters:quality(82)/media/images/e1e6912f-d509-403f-bdbe-c5cf694c3eb2.webp)
seedance-2.5/reference-to-video
Reference to Video
/filters:quality(82)/media/images/1787546478975244978_9IJS1aku.webp)
wan-3.0-prime/reference-to-video
Reference to Video
/filters:quality(82)/media/images/e5486405-20b3-407f-bc99-3f3d6abdf5dc.webp)
seedance-2.5-flash/reference-to-video-turbo
Reference to Video
/filters:quality(82)/media/images/9e27da0a-ff6d-4386-89a3-fdca16747369.webp)
seedance-2.5-flash/reference-to-video
Reference to Video
/filters:quality(82)/media/images/1785740347734124119_3h5foxHR.webp)
wan-3.0/reference-to-video
Reference to Video
Vidu Q4 Reference-to-Video generates a video of 3 to 16 seconds, up to 4K, from a text prompt and up to 15 reference images, so people, products and scenes stay consistent. Add up to 3 voice clips to give characters their own voice.
| Parameter | Required | Description |
|---|---|---|
| prompt | Yes | The scene: who does what, the camera movement and the mood. |
| images | Yes | 1 to 15 reference images, as an array of public URLs (PNG, JPEG or WebP) or Base64-encoded images. |
| audios | No | Up to 3 voice reference clips, as an array of MP3 URLs. Each clip lasts 3 to 12 seconds and is at most 50 MB. |
| aspect_ratio | No | 16:9, 9:16, 1:1, 3:4 or 4:3. Default: 16:9. |
| resolution | No | 540p, 720p, 1080p, 2k or 4k. Default: 720p. |
| duration | No | Length in whole seconds, from 3 to 16. Default: 5. |
| generate_audio | No | Generate sound with the video. Default: true. |
| seed | No | Fix the seed to compare variations. Omit it, or use -1, for a random seed. |
images and audios are arrays even when you send a single item.
@1 is the first image, @2 the second, and so on.audios.Pricing is based on the selected resolution and video duration. The price is charged per second of video, so it grows in proportion to duration (whole seconds, 3 to 16).
| Resolution | Price per second (duration = 5 seconds) | 3-second video (duration = 3) | 5-second video (duration = 5) | 10-second video (duration = 10) | 16-second video (duration = 16) |
|---|---|---|---|---|---|
| 540p | $0.045 | $0.135 | $0.225 | $0.45 | $0.72 |
| 720p | $0.095 | $0.285 | $0.475 | $0.95 | $1.52 |
| 1080p | $0.12 | $0.36 | $0.60 | $1.20 | $1.92 |
| 2K | $0.19 | $0.57 | $0.95 | $1.90 | $3.04 |
| 4K | $0.39 | $1.17 | $1.95 | $3.90 | $6.24 |
The total is the per-second rate for the selected resolution multiplied by duration. The default of 720p and 5 seconds (default settings, per run) costs $0.475. Audio generation and the other settings do not change the price.
@1, @2.audios gives the model voices to follow, and generate_audio decides whether the video has sound at all.seed makes variations easier to compare, but identical output is not guaranteed.Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/vidu/q4/reference-to-video with your input as JSON. The endpoint returns a prediction id. Start polling the result endpoint around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. On completed, read output values from data.outputs. Examples for Q4 Reference To Video below.
set -euo pipefail
: "${WAVESPEED_API_KEY:?Set WAVESPEED_API_KEY}"
REQUEST_BODY=$(cat <<'JSON'
{
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"images": [
"https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
],
"aspect_ratio": "16:9",
"resolution": "720p",
"duration": 5,
"generate_audio": true
}
JSON
)
# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
-X POST "https://api.wavespeed.ai/api/v3/vidu/q4/reference-to-video" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $WAVESPEED_API_KEY" \
-d "$REQUEST_BODY")
TASK=$(printf '%s' "$SUBMIT_RESPONSE" | jq 'if has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "$TASK" | jq -r '.id')
if [ -z "$PREDICTION_ID" ] || [ "$PREDICTION_ID" = "null" ]; then
printf 'Submission response did not contain a prediction id
' >&2
exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/$PREDICTION_ID/result"
# 2. Poll until the prediction finishes.
while true; do
RESPONSE=$(curl --silent --show-error --fail-with-body "$RESULT_URL" \
-H "Authorization: Bearer $WAVESPEED_API_KEY")
RESULT=$(printf '%s' "$RESPONSE" | jq 'if has("data") then .data else . end')
STATUS=$(printf '%s' "$RESULT" | jq -r '.status')
case "$STATUS" in
completed) printf '%s\n' "$RESULT" | jq '.outputs'; break ;;
failed|cancelled|timeout|deleted) printf '%s\n' "$RESULT" | jq . >&2; exit 1 ;;
*) sleep 2 ;;
esac
doneconst submitUrl = "https://api.wavespeed.ai/api/v3/vidu/q4/reference-to-video";
const apiKey = process.env.WAVESPEED_API_KEY;
if (!apiKey) throw new Error('Set WAVESPEED_API_KEY');
async function requestJson(url, options = {}) {
const response = await fetch(url, options);
if (!response.ok) throw new Error(await response.text());
return response.json();
}
// 1. Submit the prediction.
const body = await requestJson(submitUrl, {
method: "POST",
headers: {
"Authorization": `Bearer ${apiKey}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"images": [
"https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
],
"aspect_ratio": "16:9",
"resolution": "720p",
"duration": 5,
"generate_audio": true
}),
});
const task = body.data ?? body;
if (!task.id) throw new Error("Submission response did not contain a prediction id");
const resultUrl = `https://api.wavespeed.ai/api/v3/predictions/${task.id}/result`;
// 2. Poll until the prediction finishes.
while (true) {
const resultBody = await requestJson(resultUrl, {
headers: { "Authorization": `Bearer ${apiKey}` },
});
const result = resultBody.data ?? resultBody;
if (result.status === "completed") {
console.log(result.outputs);
break;
}
if (["failed", "cancelled", "timeout", "deleted"].includes(result.status)) throw new Error(JSON.stringify(result));
await new Promise(resolve => setTimeout(resolve, 2000));
}import json
import os
import time
from urllib.request import Request, urlopen
api_key = os.environ["WAVESPEED_API_KEY"]
headers = {"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"}
payload = {
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"images": [
"https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
],
"aspect_ratio": "16:9",
"resolution": "720p",
"duration": 5,
"generate_audio": True
}
def request_json(url, data=None):
request = Request(url, data=data, headers=headers, method="POST" if data else "GET")
with urlopen(request) as response:
return json.load(response)
# 1. Submit the prediction.
body = request_json("https://api.wavespeed.ai/api/v3/vidu/q4/reference-to-video", json.dumps(payload).encode())
task = body.get("data", body)
if not task.get("id"):
raise RuntimeError("Submission response did not contain a prediction id")
result_url = f"https://api.wavespeed.ai/api/v3/predictions/{task['id']}/result"
# 2. Poll until the prediction finishes.
while True:
result_body = request_json(result_url)
result = result_body.get("data", result_body)
status = result.get("status")
if status == "completed":
print(result.get("outputs", []))
break
if status in {"failed", "cancelled", "timeout", "deleted"}:
raise RuntimeError(result)
time.sleep(2)Q4 Reference To Video is a Vidu model for reference-guided video generation, exposed as a REST API on WaveSpeedAI. Vidu Q4 Reference-to-Video generates subject-consistent AI videos from text prompts, 1-15 reference images, and up to 3 reference audio clips, supporting character consistency, product videos, social content, brand assets, and reference-guided storytelling workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing. You can call it programmatically or try it from the playground above.
POST your input parameters to the model's REST endpoint (shown in the API tab of this playground) with your WaveSpeedAI API key in the Authorization header. Submission returns a prediction ID. Poll the result endpoint starting around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. The playground generates Python, JavaScript, and cURL examples for submitting requests and polling results. Full request/response shape is documented at https://wavespeed.ai/docs/docs-api/vidu/q4/reference-to-video.
Q4 Reference To Video starts at $0.095 per run or per second. That figure is the base price — the final charge scales with the parameters you set in the form (output size, length, count, references, or whatever knobs this model exposes), so a higher-quality or larger output costs more than a minimal one. The exact cost for your current input is shown live next to the Generate button before you submit, and the actual per-call charge is recorded on the prediction afterwards.
Key inputs: `prompt`, `images`, `aspect_ratio`, `resolution`, `duration`, `seed`. The full JSON schema (types, defaults, allowed values) is rendered above the Generate button and mirrored in the API reference at https://wavespeed.ai/docs/docs-api/vidu/q4/reference-to-video.
Reported generation time on WaveSpeedAI is around 99 seconds per request. This is an estimate, not a latency guarantee; queue time and input settings can change the total wait. live status is visible in the prediction record.
Commercial usage rights depend on the model's license, set by its provider (Vidu). Check the provider's applicable terms and WaveSpeedAI's Terms of Service before commercial use.