Image-to-video API — animate a still frame

Send a reference image and a motion prompt; get a clip that starts from that frame. Used for product turntables, avatar animation and bringing a hero image to life.

Crazyrouter is an AI API gateway that serves 6 image-to-video models — Kling, Seedance, Veo 3.1 and Hailuo — through one asynchronous endpoint that accepts a reference frame, billed per clip from $0.0357.

models

$0.0357

from, per generation

15%

max below list

Try Kling 2.6 free Get an API key

Price per clip, cheapest first

Clips from these models

Generated through this API, 2026-09-28

Kling 2.6

$0.1785 · 1.4 min · 5s

Kling 3.0

$0.357 · 2.4 min · 5s

Image-to-video models, by price per clip

Model	Vendor	Crazyrouter price (per generation)	Discount
Kling 2.6 aigc-video-kling-2.6	Kling AI	$0.0357–$0.268 / second	−15%	Try
Kling 3.0 aigc-video-kling-3.0	Kling AI	$0.0714–$0.143 / second	−15%	Try
Hailuo H3 hailuo-h3	Hailuo (MiniMax)	$0.0714–$0.251 / second	—	Try
Google Veo 3.1 aigc-video-gv-3.1	Google	$0.20–$0.60 / second	—	Try
Seedance 1.5 Pro doubao-seedance-1-5-pro	ByteDance	$1.14–$2.29 / 1M tokens	—	Try
Seedance 2.0 doubao-seedance-2-0	ByteDance	$4.00–$6.57 / 1M tokens	—	Try

Prices come from the live billing table and already include the Crazyrouter discount; list prices are each vendor's public rate, for comparison only.

What is a image-to-video API?

Image-to-video keeps the first frame fixed to your image and generates the motion described in the prompt. It gives far more control over subject identity than text-to-video, which is why e-commerce and avatar products use it.

How to choose

Product turntables

Kling 3.0 / 2.6 hold product geometry best across the rotation.

People and avatars

Seedance 2.0 and Hailuo H3 for natural facial motion.

With sound

Veo 3.1 adds synchronized audio to the animated clip.

Drafting

Seedance 1.5 Pro for cheap iterations on the motion prompt.

Animate a frame

Swap model for any row in the table; the example uses aigc-video-kling-2.6.

cURL

curl https://api.crazyrouter.com/v1/video/generations \
  -H "Authorization: Bearer $CRAZYROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"aigc-video-kling-2.6","prompt":"the camera slowly orbits the product","image":"https://example.com/frame.jpg","duration":5}'

FAQ

How do I pass the reference image?

Add an image field (URL or base64) to the same POST /v1/video/generations body; the model treats it as the first frame.

Can I control the end frame too?

Kling supports first- and last-frame inputs; see the model page for the exact fields.

Is it billed differently from text-to-video?

No — per clip at the rate in the table.

Which image-to-video model is cheapest?

Kling 2.6 at $0.0357 per clip as of the latest sync.

What image size should I send?

Match the output aspect ratio (16:9 or 9:16) and at least 1280 px on the long side for best results.