Professional Seedance V2.0 Reference-to-Video API (Multimodal Reference Generation)
ByteDance Seedance V2.0 Reference-to-Video API provides reference-guided video generation for developers and creative
teams on Best Image AI. The current API integration accepts a text prompt with at least one reference image or video, plus
optional additional image, video, and audio references. It supports configurable duration, multiple aspect ratios,
several resolution tiers, and an optional generated-sound setting for controlled video workflows.
Key Features of Seedance V2.0 Reference-to-Video API
Image and Video Reference Input: Upload multiple reference images or video clips to guide the generated result. At
least one image or video reference is required for each request.
Prompt-Directed Audio References: Add supported audio clips alongside the required visual reference media, then
use prompt instructions to ask the model to reference them for a desired role such as voice, music, timing, or
atmosphere.
Prompt-Based Reference Mentions: Refer to uploaded image, video, and audio assets from the prompt to describe how
they should influence the requested video.
Optional Generated Sound: Enable or disable generated sound through the API's supported sound setting.
Multiple Resolution Tiers: Choose from 480p, 720p, 1080p, and 4K output options exposed by the current Best Image AI
model configuration.
Flexible Duration Control: Generate videos from 4 to 15 seconds with a configurable output duration.
Flexible Aspect Ratio Support: Create landscape, portrait, square, ultrawide, and standard-format video through
the available aspect ratio options.
How to Use Seedance V2.0 Reference-to-Video API on Best Image AI
Input: A natural-language prompt with at least one reference image or video. Additional images, videos, and
supported audio clips are optional.
Reference Media: Upload multiple image and video references, plus optional audio references. Reference video and
audio clips must remain within the supported input-duration range.
Output: A generated video returned through Seedance V2.0 Reference-to-Video API integration, with generated sound
available when enabled.
Duration: 4 to 15 seconds, configurable per request.
Resolution: 480p, 720p, 1080p, or 4K.
Aspect Ratios: Supports multiple landscape, portrait, square, ultrawide, and standard formats.
Best Use Cases for Seedance V2.0 Reference-to-Video API Integration
Reference-Guided Brand Content: Use product, character, spokesperson, or scene references to guide campaign and
social video requests.
Character & IP Workflows: Supply recurring visual references when requesting new actions, environments, and
short-form scenes.
Product Video Concepts: Combine product images or clips with prompts to request demonstrations, lifestyle scenes,
and promotional variations.
Audio-Informed Creative Requests: Mention uploaded audio in the prompt and describe whether it should guide voice,
music, timing, atmosphere, or another creative element.
Automated Video Pipelines: Integrate reference uploads, prompt mentions, output settings, and generated sound into
scalable application workflows.
Note At least one reference image or video is required; audio alone cannot be used as the only reference input.
Please ensure your prompts and reference media comply with ByteDance's content safety guidelines.
Seedance V2.0 Reference-to-Video vs Competitors: Comparative Analysis
Seedance V2.0 Reference-to-Video vs. Seedance V2.0 Text-to-Video Seedance V2.0 Text-to-Video works from a text
prompt. Reference-to-Video adds supported image, video, and optional audio inputs plus prompt-based media mentions.
Seedance V2.0 Reference-to-Video vs. Seedance V2.0 Fast Reference-to-Video Both variants expose the same
reference-media types and core controls on Best Image AI. The standard variant adds 1080p and 4K resolution options, while
the Fast variant focuses on 480p and 720p workflows.
Seedance V2.0 Reference-to-Video vs. Wan 2.7 Reference-to-Video Both APIs accept image and video references. Wan
2.7 additionally exposes negative-prompt and seed controls, while Seedance V2.0 supports multiple optional audio
references and a generated-sound toggle.
Seedance V2.0 Reference-to-Video vs. Vidu Q3 Reference-to-Video Vidu Q3 Reference-to-Video uses image references
in the current Best Image AI configuration. Seedance V2.0 also accepts reference videos and optional reference audio.
Seedance V2.0 Reference-to-Video vs. Runway Video Tools Runway offers a broader interactive creation suite.
Seedance V2.0 Reference-to-Video provides a focused API workflow around prompts, supported reference uploads, output
settings, and generated sound.
Use the Seedance 2.0 Reference to Video Model Directly in Powerful AI Tools
Explore several AI creation tools for quick image and video workflows in your browser, then scale successful ideas with Best Image AI's production-ready model APIs. Best Image AI provides a unified API layer for all models, making it easy to use and scale your workflows.
// Step 1: Submit generation requestconst response = awaitfetch('https://api.flaq.ai/api/v1/video/task', {
method: 'POST',
headers: {
'Content-Type': 'application/json',
'Authorization': 'Bearer YOUR_API_KEY'
},
body: JSON.stringify({
model_name: 'seedance-v2.0-reference-to-video',
prompt: 'Use image one for the subject, video one for the movement, and audio one for the atmosphere',
resolution: '1080p',
duration: 8,
aspect_ratio: '16:9',
sound: true,
images: ['https://example.com/subject-reference.jpg'],
videos: ['https://example.com/motion-reference.mp4'],
audios: ['https://example.com/atmosphere-reference.mp3']
})
});
const { data } = await response.json();
const taskId = data.task_id;
// Use the @ (AT) reference feature in prompt through <<<...>>> placeholders.// Placeholder numbering is 1-based for each media array:// <<<image_1>>> = images[0], <<<image_2>>> = images[1]// <<<video_1>>> = videos[0], <<<audio_1>>> = audios[0]const mediaReferenceResponse = awaitfetch('https://api.flaq.ai/api/v1/video/task', {
method: 'POST',
headers: {
'Content-Type': 'application/json',
'Authorization': 'Bearer YOUR_API_KEY'
},
body: JSON.stringify({
model_name: 'seedance-v2.0-reference-to-video',
prompt: 'Place the explorer from <<<image_1>>> in the environment from <<<image_2>>>, following the camera movement in <<<video_1>>> and speaking with the reference voice from <<<audio_1>>>',
resolution: '1080p',
duration: 10,
aspect_ratio: '16:9',
sound: true,
images: [
'https://example.com/explorer-reference.jpg',
'https://example.com/environment-reference.jpg'
],
videos: ['https://example.com/camera-movement-reference.mp4'],
audios: ['https://example.com/voice-reference.mp3']
})
});
const { data: mediaReferenceData } = await mediaReferenceResponse.json();
const mediaReferenceTaskId = mediaReferenceData.task_id;
# Step 1: Submit generation request
curl -X POST https://api.flaq.ai/api/v1/video/task \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model_name": "seedance-v2.0-reference-to-video",
"prompt": "Use image one for the subject, video one for the movement, and audio one for the atmosphere",
"resolution": "1080p",
"duration": 8,
"aspect_ratio": "16:9",
"sound": true,
"images": ["https://example.com/subject-reference.jpg"],
"videos": ["https://example.com/motion-reference.mp4"],
"audios": ["https://example.com/atmosphere-reference.mp3"]
}'# Use the @ (AT) reference feature in prompt through <<<...>>> placeholders.# Placeholder numbering is 1-based for each media array:
curl -X POST https://api.flaq.ai/api/v1/video/task \
-H \
-H \
-d
0
url
break
if
'failed'
console
error
data
task_status_msg
break
await
new
Promise
resolve =>
setTimeout
10000
'seedance-v2.0-reference-to-video'
'prompt'
'Use image one for the subject, video one for the movement, and audio one for the atmosphere'
'resolution'
'1080p'
'duration'
8
'aspect_ratio'
'16:9'
'sound'
True
'images'
'https://example.com/subject-reference.jpg'
'videos'
'https://example.com/motion-reference.mp4'
'audios'
'https://example.com/atmosphere-reference.mp3'
'data'
'task_id'
# Use the @ (AT) reference feature in prompt through <<<...>>> placeholders.
# Placeholder numbering is 1-based for each media array:
'Place the explorer from <<<image_1>>> in the environment from <<<image_2>>>, following the camera movement in <<<video_1>>> and speaking with the reference voice from <<<audio_1>>>'
'{
"model_name": "seedance-v2.0-reference-to-video",
"prompt": "Place the explorer from <<<image_1>>> in the environment from <<<image_2>>>, following the camera movement in <<<video_1>>> and speaking with the reference voice from <<<audio_1>>>",
"resolution": "1080p",
"duration": 10,
"aspect_ratio": "16:9",
"sound": true,
"images": [
"https://example.com/explorer-reference.jpg",
"https://example.com/environment-reference.jpg"
],
"videos": ["https://example.com/camera-movement-reference.mp4"],
"audios": ["https://example.com/voice-reference.mp3"]
}'
Polling Example
# Step 2: Poll for results# Replace {task_id} with the task_id returned from the submit response
curl -X GET "https://api.flaq.ai/api/v1/video/{task_id}" \
-H "Authorization: Bearer YOUR_API_KEY"
