Reference to Video

Reference Images (optional)(0/9)

Upload at least one image or video (audio-only is not allowed). In the prompt, tag assets with @image1, @video1, @audio1.

Reference Videos (optional)(0/3)
Reference Audios (optional)(0/3)
0 / 5000
Cost 14 creditsRemaining 0 credits
Video Preview
Example

Ready for Generating

What is Reference to Video?

Reference to video lets you direct a clip with several inputs at once. Use images for characters, outfits and scenes, video clips for motion and camera work, and audio for rhythm or voice, then tell the model what each one is for. It is the most precise way to keep characters and style consistent across shots.

Images, video and audio together

Combine reference images, short video clips and audio files in one generation, depending on the model.

Tag references in the prompt

Refer to each input as @image1, @video1 or @audio1 so the model knows exactly what to use and how.

Consistent characters and style

Lock a character’s face, clothing or a visual style with reference images and reuse them across clips.

Models

Models for Reference to Video

Every model below works in this tool. Prices are for the default settings and update with the generator; higher resolution or longer videos cost more.

Seedance 2.5 Flash

66 credits
with default settings
Up to 10 reference images

Faster 480p / 720p route. 6.138 / 13.2 credits/s with no video input. Default 5s 720p ≈ 66 credits. Supports text-to-video, first/last-frame image-to-video, reference-to-video, video edit, and video extend.

Seedance 2.5 Flash Edit

65 credits
with default settings
Up to 10 reference images

Faster 480p / 720p route. 6.138 / 13.2 credits/s with no video input. Default 5s 720p ≈ 66 credits. Supports text-to-video, first/last-frame image-to-video, reference-to-video, video edit, and video extend.

Seedance 2.5 Flash Extend

66 credits
with default settings
Up to 10 reference images

Faster 480p / 720p route. 6.138 / 13.2 credits/s with no video input. Default 5s 720p ≈ 66 credits. Supports text-to-video, first/last-frame image-to-video, reference-to-video, video edit, and video extend.

Grok Imagine 1.5

23 credits
with default settings
Up to 7 reference images

One model for text-to-video (0 images), image-to-video (1 image), and reference-to-video (2–7 images). 3.808 / 6.664 / 11.9 credits/s at 480p / 720p / 1080p, plus 0.476 credits per input image. Default 6s 480p ≈ 23 credits.

HappyHorse 1.1

42 credits
with default settings
Up to 9 reference images

Physically realistic motion. Text-to-video, image-to-video (first frame), and reference-to-video (1–9 images via character1…). 8.3935 / 11.2473 credits/s at 720p / 1080p. Default 5s 720p ≈ 42 credits.

Seedance 2.5

101 credits
with default settings
Up to 30 reference images

9.34 / 20.09 / 36.16 credits/s at 480p / 720p / 1080p. Default 5s 720p ≈ 101 credits. Supports 4–30s (or Auto), web search, optional MOV output. With video references, lower rates apply and reference duration is billed too.

Minimax H3

21 credits
with default settings
Up to 9 reference images

Text-to-video, image-to-video (first/last frame), and reference-to-video. 5.168 credits/s at 768p, 8.398 credits/s at 2K (same for all modes). Default 4s ≈ 21 credits (768p) or ≈ 34 credits (2K). Reference videos add their duration to billing.

Seedance 2.0 Mini

14 credits
with default settings
Up to 9 reference images

Lightweight drafts. 1.2555 / 2.7 credits/s at 480p / 720p. Default 5s 720p ≈ 14 credits. Video references use lower rates and bill input + output seconds.

Seedance 2.0 Fast

55 credits
with default settings
Up to 9 reference images

Faster Seedance route (~20% cheaper). ≈ 5.08 / 10.99 credits/s at 480p / 720p. Default 5s 720p ≈ 55 credits.

Seedance 2.0

68 credits
with default settings
Up to 9 reference images

6.2775 / 13.5 / 33.75 credits/s at 480p / 720p / 1080p. Default 5s 720p ≈ 68 credits. With video references, lower per-second rates apply and reference duration is billed too.

Gemini Omni 1.2 Flash

23 credits
with default settings
Up to 10 reference images

Google's multimodal video model with native audio always on: 3 to 15-second clips from 360p for quick drafts up to 4K. Image to video takes a first frame or first and last frames; reference to video combines up to 10 images and 3 videos.

How it works

How to Create a Video from References

Four steps to direct a multimodal clip:

1

Upload references

Add images, video clips and audio. Seedance models accept all three; some models accept images only.

2

Tag them in your prompt

Write the scene and point to each file, for example "the woman from @image1 dances like @video1, with @audio1 as music".

3

Choose model and length

Pick a model, aspect ratio and duration.

4

Generate and review

Check how each reference was used, then adjust the tags or prompt and generate again.

Examples

Reference to Video prompt ideas

Pick one to fill the generator above, then adjust it to your idea.

Dance with a reference move

Product with a camera move

Two characters together

Consistent outfit

Scene from a photo

Music video

Reference to Video FAQ







Direct Your Video with References

Combine images, motion and sound, and let the model bring them together.

Reference to Video AI Generator | Multimodal Video | SeedDream