Reference to Video

Model
PROMPT
INPUT
Reference materials*

Upload limits: 9 images (JPG / PNG / WebP / HEIC, 30MB each) · 3 videos (MP4 / MOV, 2-15s and 64MB each, 15s total) · 3 audio (MP3 / WAV, 2-15s and 15MB each, 15s total)

Images0 / 9
Videos0 / 3
Audio0 / 3
ASPECT RATIO
VIDEO LENGTH
RESOLUTION
Output Number
Generate several videos at once
Generate Audio
Add natural-sounding audio
Total Credits:
20 Credits

Sign in with Google in 1 second — check in daily for free credits (up to 60 a week)

AI-Powered

Bring Your Characters into New Scenes with AI

Upload reference images of your characters or products and let AI create new scene videos while keeping their appearance consistent.

Key Features

  • Upload multiple reference images (supported models only)
  • Keep the look of characters and products in new scenes
  • Great for episodic short videos and ad creatives

Powered by Leading AI Models

MiniMax H3
MiniMax
Seedance 2.5
ByteDance

How reference-to-video works

Upload reference images of the character or product you want to keep consistent, describe the new scene you want, and generate. The model reproduces the look of your reference material in a scene that never existed.

This is the mechanism behind the AI video templates on this site, and it is what makes a series of clips look like the same character or product every time.

When to use it

Reach for reference-to-video when appearance consistency matters more than a single good frame.

  • Episodic short videos where the same character appears in every episode
  • Running a product through several different scenes and settings
  • Reproducing a movement or composition from a reference clip

FAQ

What is the difference between image-to-video and reference-to-video?

Image-to-video animates one image — the output starts from that frame. Reference-to-video treats your images as reference material and generates a new scene that keeps the subject's appearance, so you can place the same character or product somewhere else entirely.

How many reference images can I use?

It depends on the model. Multiple reference images are supported on supported models; the upload panel shows how many slots the selected model accepts.

Is it free to use?

You get free credits via daily check-in, and no credit card is required. Credit costs are shown before you generate.

Can I use reference-to-video clips commercially, and is there a watermark?

Videos generated on any paid plan (monthly/yearly subscription or credit pack) download without a watermark and can be used commercially within our Terms of Service. Browser playback is always watermark-free; free-credit downloads carry a Sesshu watermark and are limited to personal, non-commercial use.

Which AI models support reference-to-video?

You can choose between models such as Seedance 1.5 Pro, Seedance 2.0 series, Vidu Q4, and Wan depending on your scene and budget. Each model’s reference slot limit and exact credit cost are shown on screen before you generate.

Can I write prompts in my own language and use it on mobile?

Yes. Sesshu runs directly in desktop and mobile browsers with no app download required, supports 32 interface languages, and accepts prompts written in your own language.

Ready to create amazing videos? Sign up now and get started!