VideoSpicy route · no extra filterEarly access

首尾帧生成

Start here. End there.

Two stills define the shot: where it opens and where it lands. The model fills the motion in between, so a turn, a walk to the window, or a reveal happens exactly as you framed it.

首帧定开场，尾帧定落点，中间的运动交给模型。

Open in Chat All tools

Price From $0.051 per second on the cheapest model below.

First & last frame video

from $0.051/sec

Drop a start frame and an end frame hereor click to choose filesFiles stay in your project. Nothing is trained on.

length

frame

Invite code opens this tool in Chat. No code yet? Join the waitlist — we open the most-requested tools first. Open in Chat

How it works

Three moves.

Frame the two ends

Same character, same lighting; change pose, position or expression.

Describe the path

One motion: “walks to the window and turns back.”

Run

If the path wanders, shorten the duration.

Illustrative sample from MiniMax H3 Image-to-Video · prompt: “The warrior raised the sword in his hand high and brandished it.”

You give

Inputs.

Start frame
End frame (same character, same room)
Motion prompt
Duration

You get

Outputs.

Clip that lands on your end frame

Under the hood

Models behind this tool.

Each model lists its own price and content policy. The spicy route is MiniMax H3; other models follow their vendor’s rules.

FAQ

Short answers.

Do both frames need the same aspect ratio?

Yes. Crop both to the target ratio before uploading.

How different can the frames be?

Change pose, position or expression — not the person, the outfit or the room.

Which models support this?

MiniMax H3 image-to-video takes an optional last frame; Kling, Vidu and Wan have dedicated start-end modes listed below.

How do I get access?

Enter an invite code to open the tool in Chat, or join the waitlist from this page. We open seats in batches and prioritise the tools people ask for most.

Related tools