UnlimitedVideo

PicassoIA Reference To Video API

Reference to Video

Reference-guided video with audio, clips of 1 to 15 seconds and 480p or 720p output.

1
2
3
4curl -s -X POST https://api.picassoia.com/v1/models/picassoia/picassoia-reference-to-video/predictions \
  -H "Authorization: Bearer $PICASSOIA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"input":{"prompt":"Image 1, the woman with red hair, walks into Image 2, the cafe, and sits by the window. Soft cafe ambience.","images":["https://example.com/woman.jpg","https://example.com/cafe.jpg"],"resolution":"720p"}}'

Input

PromptRequired

ImagesOptional

Add Images (Up to 9)

Paste a public https URL, or pick one of ours. From your code you can also send a data URL.

Or Pick One of Ours

VideoOptional

Add a Video

Paste a public https URL. The API takes a video or an audio as a URL only, not as a data URL.

Audio FileOptional

Add an Audio File

Paste a public https URL. The API takes a video or an audio as a URL only, not as a data URL.

Audio

Keep the generated audio track.

Duration5

115

The length of the clip, in seconds.

Generate

Opens PicassoIA Reference To Video in Picasso IA with this prompt

Result

No examples for this model yet.

Price

$0 per request with the Infinite plan, whatever the settings.

More Models by PicassoIA

Other models you can call with the same API key, the same request shape and the same price.

Model	Modes	Limits	Price	Added
PicassoIA Reference To VideoThis Model	Reference to Video	audio and clips of 1 to 15 seconds	$0 / run	November 10, 2026
Seedance 2.5 Lite	Text and Image to Video	audio and clips of 5, 10 or 15 seconds	$0 / run	August 10, 2026
PicassoIA Video	Text and Image to Video	audio and clips of 1 to 20 seconds	$0 / run	June 9, 2026
PicassoIA Image	Text to Image	1 or 2 images per request and 7 aspect ratios	$0 / run	May 25, 2026
PicassoIA Image Editor Pro	Image Editing	up to 4 input images and 1 or 2 images per request	$0 / run	May 25, 2026

Call PicassoIA Reference To Video from Your Code

Create an API Key

Create a key in the API Keys section of your account and keep it on your server, in the PICASSOIA_API_KEY environment variable. Creating keys and predictions needs an active Infinite plan.

Send the Request

Send a POST request to /v1/models/picassoia/picassoia-reference-to-video/predictions with your API key in the Authorization header and your settings in the JSON body's input object. The API answers 201, usually within seconds, with the prediction and an estimate of when to poll. It never waits for the result.

Poll and Download

Poll the prediction's urls.get, waiting eta.next_poll_in_seconds between calls, until its status is succeeded, failed or canceled. If it succeeded, download the MP4 video. The Node.js and Python code at the top of this page sends the request and polls the prediction for you.

PicassoIA Reference To Video API FAQ

What does the PicassoIA Reference To Video API do?

PicassoIA Reference To Video is a model for reference-guided video, with audio, clips of 1 to 15 seconds, 480p or 720p output and up to 9 input images. Through the PicassoIA API, you create a prediction with one JSON request from your server, poll it and download the result when it is ready.

How do I call the PicassoIA Reference To Video API from my code?

Send a POST request to /v1/models/picassoia/picassoia-reference-to-video/predictions with your API key and the input, then poll the prediction until its status is succeeded, failed or canceled. The code card at the top of this page writes the request for the settings you choose.

What are the limits of PicassoIA Reference To Video?

It supports audio, clips of 1 to 15 seconds, 480p or 720p output and up to 9 input images. Its API reference lists the limits of each field. The API refuses anything outside them with 422 invalid_input and names every field at fault. It refuses unknown fields too, so a typo never falls back to a default.

How much does the PicassoIA Reference To Video API cost?

$0 per request with an active Infinite plan: API predictions use no credits. Without the plan, creating a prediction answers 403 plan_required.

How many PicassoIA Reference To Video predictions can run at once?

Up to 5 predictions per account can be queued or running at the same time, whatever the model. The limit is shared by your API keys, the generations you run through MCP (ChatGPT, Claude), Supercomputer and the ones included in your plan running on picassoia.com. Creating one more through the API answers 429 with a Retry-After header. On picassoia.com the same count stops at 4.

What does PicassoIA Reference To Video return?

When the prediction succeeds, its output is the URL of an MP4 video, hosted by Picasso IA. Download what you want to keep.

A young skateboarder in a black t-shirt and baggy jeans performs tricks at a brutalist concrete skatepark on a sunny afternoon. The camera alternates between a fisheye lens ground-level POV following his board and wider angles capturing kickflips and grinds. Harsh midday shadows, raw street-skating documentary style, warm golden tones.





		hasta 4 imágenes de entrada y 1 o 2 imágenes por peticiónup to 4 input images and 1 or 2 images per requestما يصل إلى 4 صور مدخلة وصورة أو صورتين لكل طلبaté 4 imagens de entrada e 1 ou 2 imagens por requisiçãojusqu'à 4 images d'entrée et 1 ou 2 images par requêteअधिकतम 4 इनपुट इमेज और हर रिक्वेस्ट पर 1 या 2 इमेज