Image to Video AI API
Newest-generation video models that turn a still into a coherent, realistic clip, with native audio on Motion 3, up to 1080p and 30 seconds. One REST call, from $0.05 per second.
{
"status": "succeeded",
"output": {"video_url":"…/gen_8f3k2m.mp4"},
"cost_usd": 1
}How the AI animates your image
curl https://api.spicyapi.com/v1/videos/generations \ -H "Authorization: Bearer $SPICYAPI_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "spicy-motion-3", "prompt": "slow push-in, she turns toward the camera", "image_url": "<a URL from /v1/images/generations>", "resolution": "720P", "duration": 5 }'# 202 returns a task id, then poll it or take the webhook:curl https://api.spicyapi.com/v1/videos/tasks/<id> \ -H "Authorization: Bearer $SPICYAPI_KEY"
The models behind the API
Generated with the API
Raw outputs, no retouching. Blurred until you confirm you are 18 or older.
Next-generation video models
Same person, every clip
Give Motion 3 up to 10 reference images or a saved character and it keeps that person consistent in a new clip.
Dialogue, music, SFX
Motion 3 renders sound with the picture, no separate audio pipeline. Dialogue written in the prompt is spoken.
Two ways in
Animate a generated still, or go text-to-video with no source at all.
Motion 3 Fast
Same inputs and quality tier on an accelerated backbone, for a noticeably faster turnaround.
Images, video and chat
The same Bearer key and billing for every model.
Transparent pricing
Pay per second at the resolution's rate. Failed jobs are refunded, and duration -1 is settled to the length actually rendered.
Unit prices, per second
Prepaid credits in USD, spent at each model's rate when a job succeeds. Failed or blocked generations are never billed and the balance never expires. Top up from $50 by card or crypto.
Common questions
Which AI model is best for image to video?
Motion 3 for the highest quality and native audio, Motion 3 Fast when turnaround matters, and Motion 3 with an action (one of the ready-made sex acts from GET /v1/videos/actions) for sex scenes with no motion prompt at all.
Does it generate audio?
Yes on Motion 3 and Motion 3 Fast: dialogue, music and sound effects are rendered with the video. Without a first frame you can also pass reference audio (up to 5 clips, 15s in total) to guide voice, tone and beat. On Motion 2, a driving audio_url becomes the soundtrack and the mouth follows it.
Can I do text to video without an image?
Yes. Motion 3, Motion 3 Fast and Spicy Video 1 accept a prompt alone, with an aspect_ratio (16:9, 9:16, 1:1, 4:3, 3:4, 21:9 on Motion 3).
What does it cost?
From $0.10 per second at 480p. 720p and 1080p cost more per second; see the pricing page for every rate.
Is it OpenAI SDK compatible?
Images and chat are. Video uses the same base URL and Bearer key with a simple task-polling flow.
Experience the future of AI video generation
Top up from $50, create a key and render your first AI video in minutes.