text→video · img→videoVideo

sora-2-openai-12s

sora-2-openai-12s

Sora 2 text/img-to-video, 12 seconds.

Input
Prompt *
Images0/1 · optional
Up to 1 reference images (JPG / PNG)
Aspect ratio
Duration
Count
Running a model uses your prepaid balance, so it needs an account. Top up
OutputIdle

Run the model and the result shows up here.

Model introduction

Overview

Sora 2 text/img-to-video, 12 seconds.

Key capabilities

Text-to-video and image-to-video

Prompt alone drives text-to-video; add reference images to drive the shot from a picture.

Selectable duration

Each model has its own legal duration set; only accepted values are ever submitted.

Fixed ratios and resolutions

Ratios map to the exact upstream pixel sizes, so a job is never rejected for an illegal size.

Billed per second or per call

Charged from your prepaid balance per the model's billing mode; failures are refunded.

Parameters

ParameterDescription
Modestext→video · img→video
Size1280x720 / 720x1280 (only)
Durationfixed 12
Reference imagesup to 1, optional
Billingper second
  • Duration fixed at 12 seconds.
  • Only 1280x720 / 720x1280 are supported.
  • Optional single reference image; the gateway maps it to upstream input_reference.

Pricing

per second

Use cases

Short-form video and adsGo from a one-line script or a single key frame to usable footage.

Product demosDrive camera motion from a product still to produce demo clips fast.

Storyboard previewsTurn storyboards into moving previews before a real shoot.

Social assetsBatch out portrait or landscape assets sized for each channel.

Join our Discordsora-2-openai-12s — GenRelay