HiAPI
OverviewModels MarketAPI KeysUsage StatisticsCall LogsBillingReferralPlaygroundStorageChangelogContact UsSettings
Display unit
N
Powered by hiapi
Settings

Welcome

Contact Us

Try Seedance 2.5 reference-to-video with video, image, and audio references. Learn @video1 mappings, review 4–30 second pricing, and call POST /v1/tasks.

Provider: ByteDance

Category: video generation

Endpoint: /v1/tasks

Status: Available

Cost: See live page pricing

Back to Models

Seedance 2.5 Reference to Video

New
by ByteDanceVideo

ByteDance Seedance 2.5 reference-to-video drives generation from up to 10 reference videos totalling 30 seconds, optionally combined with up to 30 reference images and 10 audio clips, transferring look, camera language and voice into a new clip of up to 30 seconds.

Pricing

Standard Usage

242 Credits/ s

Input

0 / 20000
5s
4s1-second steps30s

Choose 4-30 seconds in 1-second steps

More parameters

Upload assets

Preview after upload
0/10

Up to 10 clips; MP4 or MOV; up to 200MB and 2-30 seconds per clip; combined duration up to 30 seconds; Items: 1 - 10

0/30

Up to 30 images; jpeg, png, webp, bmp, tiff, or gif; up to 30MB each; 300-6000 px per side and a 0.4-2.5 aspect ratio; Up to 30 items

0/10

Up to 10 clips; WAV or MP3; up to 15MB and 2-30 seconds per clip; combined duration up to 30 seconds; Up to 10 items

Generation settings

Choose the video format. MP4 is broadly compatible; MOV is intended for professional post-production

Generation preferences

Generate synchronized audio for the video

Use web search signals for generation

Estimated cost1,214 Credits242 Credits / s × 5s

Output

Ready for generation

Configure your parameters and click "Run" to see the output here.

--GPU-T4

References and results

Three workflows from motion transfer to multimodal references

See how camera language transfers to a new product, motion changes visual style, and image or audio references shape a new shot. Every example keeps its copyable original prompt.

Video + image referenceVideo-only referenceVideo + image + audio
Video + image reference4s480p9:16

Transfer a shoe commercial's camera language to a perfume bottle

The reference video supplies shot progression, rotation, macro framing, pacing, and light transitions. The image locks the amber bottle's identity while the model builds a new vertical fragrance commercial.

Original prompt

Use @video1 strictly as the reference for shot progression, product rotation, camera paths, pacing, macro close-ups, and lighting transitions. Replace the shoe completely with the perfume bottle from @image1. Preserve the bottle's exact silhouette, amber glass color, cap proportions, label placement, and premium material appearance in every shot. Present it as a luxury fragrance commercial on a dark obsidian pedestal, with warm amber rim light, controlled reflections, fine atmospheric mist, and precise macro detail. Follow the visual rhythm and camera choreography of @video1 while keeping one consistent perfume bottle throughout. Do not transfer shoe laces, soles, fabric textures, or shoe-shaped geometry into the bottle. Do not invent extra products, readable brand text, logos, hands, or people.

Input references and @token mapping

@video1Reference video10.1s · 1080 × 1920

Shot progression, rotation, macro camera, and pacing

@image1Reference image

Amber perfume bottle identity and material

Generated result

Generated video · audio track included4s · 480p · 9:16
Video-only reference4s480p16:9

Rebuild a corgi chase as a handmade stop-motion film

One video carries the corgi's stride, forward momentum, low tracking camera, and sunset rhythm. The prompt changes the visual medium while preserving a continuous performance.

Original prompt

Use @video1 as the sole visual reference. Preserve the corgi's recognizable markings, running cadence, stride timing, body orientation, forward momentum, and the low side-tracking camera movement. Reimagine the entire scene as a handcrafted stop-motion film made from felt, wool, miniature grass, and painted paper scenery. Keep the warm sunset backlight and continuous chase rhythm, with subtle handmade texture and natural depth of field. Maintain one coherent corgi throughout the shot, anatomically stable legs, smooth forward motion, and a continuous camera path. Do not introduce additional animals, text, logos, abrupt cuts, or unrelated objects.

Input references and @token mapping

@video1Reference video3.0s · 1280 × 720

Stride timing, forward motion, and low tracking camera

Generated result

Generated video · audio track included4s · 480p · 16:9
Video + image + audio4s480p16:9

Combine workshop motion, crystal-fox identity, and a sound reference

Workshop motion comes from the video, the fox sculpture's identity comes from the image, and the workshop track is submitted separately as @audio1. The result retains an audio track; this case does not claim frame-accurate beat synchronization.

Original prompt

Use @video1 for the artisan's hand movements, workshop action, camera choreography, macro framing, and the progression from raw material to the final reveal. Replace the original glass object with the crystal fox sculpture from @image1. Preserve the fox's translucent faceted geometry, pointed ears, curled tail, crystalline color, and elegant proportions throughout the sequence. Show the artisan carefully shaping the crystal fox in a warm glass workshop, with realistic tools, glowing furnace light, controlled reflections, and physically plausible hands. Use @audio1 as both the soundtrack and timing guide. Synchronize the furnace flare with the opening low pulse, important tool contacts with glass chime accents, and the final crystal-fox reveal with the last shimmering swell. Keep the fox an inanimate glass sculpture rather than turning it into a living animal. Maintain one coherent workshop, one consistent sculpture, and smooth temporal continuity. Do not add text, logos, extra sculptures, deformed hands, or unrelated scene changes.

Input references and @token mapping

@video1Reference video5.0s · 1280 × 704

Hands, workshop action, macro framing, and reveal

@image1Reference image

Crystal fox silhouette, facets, color, and proportions

@audio1Reference audio5.0s · MP3

5-second audio reference

Independent sound reference extracted from the workshop clip

Generated result

Generated video · audio track included4s · 480p · 16:9

Related video models

Choose by exact endpoint control, external motion references, or prompt-only creation.

Control endpoint frames

Seedance 2.5 Image to Video

Use a first frame or first/last frames when the shot must begin or end on exact images.

Create without references

Seedance 2.5 Text to Video

Use the text route when the composition and motion should be created entirely from a prompt.

Alternative reference workflow

HappyHorse 1.1 Reference-to-Video

Compare another reference-guided route when the project needs a different motion or style response.

Compare AI video APIs

About Seedance 2.5 Reference to Video

Seedance 2.5 Reference to Video extracts motion, performance, camera language, and rhythm from one or more videos, then combines optional image and audio guidance into a new shot. It is not a video editor or extension tool and does not lock exact endpoint frames.

The Playground and API use the same model ID: seedance-2.5/reference-to-video. Each media type is numbered independently by upload order, and the prompt assigns control through @video1, @image1, and @audio1.

Provider
ByteDance
Task
Multimodal reference-to-video
Required input
Prompt + 1–10 reference videos
Starting price
242 Credits / second

Model specifications

A reference video is required, while images and audio provide additional control. Total video references and total audio references each have separate thirty-second limits.

Required reference video
1–10 clips
Output duration
4–30s
Optional references
30 images10 audio clips
Resolution
480p720p
Aspect ratio
16:94:31:13:49:1621:9adaptive
Output and audio
MP4MOVSynchronized audio by default
Model ID
seedance-2.5/reference-to-video
API protocol
POST /v1/tasksAsync
Video
MP4 / MOV; 2–30 seconds per clip, 24–60 fps, under 200MB, and at least 409600 pixels.
Image
JPEG / PNG / WebP / BMP / TIFF / GIF; under 30MB per image.
Audio
WAV / MP3; 2–30 seconds and under 15MB per clip.

480p is 242 Credits / second; 720p is about 542 Credits / second.

How to use Seedance 2.5 Reference to Video

The site-wide three-step framework is adapted for multimodal token mapping and intentionally excludes first/last-frame logic.

@video1
@image1

Seedance 2.5 Reference to Video API quickstart

Keep model at the request root and pass at least prompt plus reference_video_urls in input. The example demonstrates all three @ references; remove optional image or audio arrays when they are unnecessary.

curl --request POST 'https://api.hiapi.ai/v1/tasks' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "seedance-2.5/reference-to-video",
  "input": {
    "prompt": "Use the camera rhythm from @video1, keep the product identity from @image1, and follow the beat of @audio1. Create one coherent commercial shot.",
    "reference_video_urls": [
      "https://example.com/reference-motion.mp4"
    ],
    "reference_image_urls": [
      "https://example.com/product-reference.jpg"
    ],
    "reference_audio_urls": [
      "https://example.com/beat-reference.mp3"
    ],
    "resolution": "480p",
    "duration": 5,
    "aspect_ratio": "adaptive",
    "generate_audio": true,
    "output_format": "mp4"
  }
}'
Replace example URLs with publicly accessible media.Open full API docs

Frequently asked questions

Why does Seedance 2.5 reference-to-video require a reference video?

The model is designed to extract motion, camera language, timing, or performance from video, so reference_video_urls must contain at least one clip. Images and audio can add identity, style, or sound guidance, but cannot replace the required video.

How is reference-to-video different from image-to-video?

Reference-to-video transfers motion, camera language, and rhythm from a clip without promising an exact first or last frame. Use seedance-2.5/image-to-video to extend one image or lock endpoint frames. Reference-to-video is not video editing or extension.

How do @video1, @image1, and @audio1 map to uploaded media?

Each media array is numbered independently in upload order. The first reference video is @video1, the first image is @image1, and the first audio clip is @audio1. State whether each item contributes motion, identity, style, or sound.

How many references does Seedance 2.5 reference-to-video accept?

It accepts one to ten reference videos, up to thirty images, and up to ten audio clips. Reference videos can total thirty seconds and audio references can separately total thirty seconds. Each video must be 2–30 seconds, 24–60 fps, and at least 409600 pixels.

How is Seedance 2.5 reference-to-video priced?

Pricing depends only on output resolution and requested duration. Reference count, reference duration, and generate_audio do not currently change the per-second tier. See the live page price and task record for the final charge.

Can Seedance 2.5 reference-to-video generate synchronized audio?

Yes. generate_audio defaults to true and can produce synchronized dialogue, effects, and background audio. Set it to false for video-only output. An audio reference can guide sound or rhythm but still requires at least one video reference.

How do I call the Seedance 2.5 reference-to-video API?

POST an asynchronous task to /v1/tasks with model at the request root set to seedance-2.5/reference-to-video. Put prompt and reference_video_urls in input, then add optional image, audio, resolution, duration, and output settings. The create request returns a taskId.

@audio1
Video required
01

Start with at least one reference video

Use video for motion, camera language, timing, or performance. Add subject images or audio only when they provide a distinct, non-conflicting role.

@image1
@video1
@audio1
02

Assign each reference a role in the prompt

Use @video1, @image1, and @audio1 in array order and state which item controls motion, identity, style, or sound instead of saying only “use these references.”

Playground · 480p
POST /v1/tasks
03

Validate reference mapping at short 480p

Check token order, identity consistency, and motion transfer first, then raise duration or resolution and create the production task with seedance-2.5/reference-to-video.