P-Video
P-Video is Pruna’s performance video generation model, delivering state-of-the-art AI video in seconds.
Designed for professional results, it offers a rare combination of speed, affordability, visual quality, draft mode for rapid iteration, and high control with text prompts and image references, making it the go-to choice when you need fast, beautiful, and reliable video generation.
Note
Some visuals shown here are inspired by, derived from brand assets, or reminiscent of representative brands across various industries and have been adapted for demo purposes. Also, when using P-Video, make sure to respect the copyright of the images you use as input and the content you generate.
Pricing:
Resolution |
Draft mode is OFF |
Draft mode is ON |
|---|---|---|
720p |
$0.02 per second of output video |
$0.005 per second of output video |
1080p |
$0.04 per second of output video |
$0.01 per second of output video |
Tip
Test it now in the P-Video Playground.
Prompt formula
Subject(s) - What/who is the focus? - Examples: “young woman with silver hair”, “knitted purple prune character”
Action(s) - What are they doing? - Examples: “running”, “dancing”, “reading a book”
Scene(s) - Where is it happening? - Examples: “urban street”, “mountain peak”, “cozy room”
[Optional] Camera Movement - How is it filmed? - Examples: “static camera”, “zooming in”, “tracking from behind”
[Optional] Lighting - What lighting conditions? - Examples: “warm natural light”, “golden hour”, “neon glow”
[Optional] Style - Visual tone/aesthetic? - Examples: “photorealistic”, “documentary style”, “cinematic”
[Optional] Audio - What audio elements? - Dialogue: “character reading aloud: ‘Your text here’” - Music: “upbeat electronic music”, “gentle background music” - Sound effects: “footsteps”, “ambient room tone”
Tip
For full prompting guidelines, see the Video Generation guide.
Key Features
P-Video delivers exceptional performance for video generation with the following key capabilities:
- Built-in Audio Generation
Native audio generation with dialogue support and lip sync. P-Video can generate synchronized audio tracks matching the visual motion and pacing of your video.
- Custom Audio Import
Import your own audio files for video generation. Supports custom audio input for music videos, voiceovers, and sound effects.
- Draft Mode
4× faster generation for rapid iteration. Draft mode enables quick preview cycles before committing to production-quality output.
- Prompt Upsampling
Ultra-fast prompt upsampling with full user control. Start with simple prompts and iteratively refine to achieve desired results.
- Aspect Ratios
Multi-aspect ratio support for flexible output configurations. Supports 16:9, 9:16, 1:1, 4:3, and 3:4 formats for different distribution channels.
A rendered image of a knight standing in a medieval village near the entrance of castle gates, detailed surroundings, natural lighting, realistic textures.
(24fps, 720p, draft mode) The knight is walking through the medieval village and enters the castle gate.
(48fps, 1080p) The knight is walking through the medieval village and enters the castle gate.
luxury product photography of a knitted purple prune action figure, elegant minimalist composition, soft studio lighting, commercial photography quality, clean background, professional product styling
Music generated using: https://replicate.com/minimax/music-1.5
(integrated audio) The character is dancing singing a poppy jazzy song "P-Video in action, right on cue, Pruna's craft, creating scenes anew, State-of-the-art, no waiting in line, AI video, done in record time. State-of-the-art, no waiting in line AI video, done in record time"
(custom audio) The character is dancing and singing a jazzy song.
A group of adults sit at a long table in a cozy Parisian living room filled with artsy furniture and metal water pitchers, viewed from the side. The atmosphere is joyful and relaxed.
(48fps, 1080p, no-audio) The group of adults is sitting at the table, chatting and laughing.
(48fps, 1080p, audio) The black man on the left of the table says: "Did you know P-Video is great at generating audio". It is quiet for a couple of second, then the rest of the people respond affirmatively.
The word "PRUNA" is made of soft, flowy pruple fur on a vibrant-colored floor, well-lit by sunlight on a bright afternoon.
(no upsampling) The letters are dancing for a commercial about "Pruna".
(upsampling) The letters are dancing for a commercial about "Pruna".
A female model standing on an improvised runway in a flowing evening gown, viewed from the front, illuminated by studio lighting, elegant and sophisticated atmosphere, capturing a sense of graceful poise
(horizontal) The model is walking down the runway, viewed from the front.
(vertical) The model is walking down the runway, viewed from the front.
Domain Use Cases
P-Video excels across various professional use cases with the following applications:
- Avatar
Create highly consistent talking avatars with strong lip sync from a single image for corporate communication, product promotion, or fictional characters.
- Social Ads
Build your own AI-powered ads studio — leveraging image consistency, multi-resolution output, and fast iteration to scale social video production effortlessly.
- Music Clips
Turn your own singing or original audio into coherent music videos with solid lip sync and visual alignment.
- Product Animated Images
Convert static or low-resolution e-commerce product images into animated videos to increase engagement and conversion.
Two French men speaking English with Man 1 (on the left) "Have you tested P Video yet? The speed-to-quality cost ratio is impressive." Man 2 (on the right) "I did! I made a 10-second video in 10 seconds... for only 10 cents. I really love it."
A 3D rendered character, rotating to show outfit from all angles, clean studio background, e-commerce style
The man say: "You! Join the fight!" and he points to the screen.
The camera pans from left to right. A heavy, deep male voice says: "Motorcycles! Faster than ever before!"
The woman in the photo moves naturally. She says: "Find the house that's perfect for you, and your family". She stays in position.
The character is singing a jazzy song.
The character is rapping along to the music.
Music generation is done using: https://replicate.com/minimax/music-1.5
subtle fashion editorial movement, slight playful movements, model maintains confident presence, natural breathing motion, gentle shoulder shift, soft micro head movement, static camera, premium ecommerce showcase, controlled minimal motion. Audio: Minimal Pop Groove
Full-body shot. Static camera and the person stands still. The person maintains confident presence, natural breathing, gentle shoulder shift, soft micro head movement, premium ecommerce showcase, controlled minimal motion. He points at the camera. Audio: Minimal Poppy Hip Hop
Full-body shot. Static camera and the person stands still. The person maintains confident presence, natural breathing, gentle shoulder shift, soft micro head movement, premium ecommerce showcase, controlled minimal motion. Person looks at the camera. Keep character features. Audio: Minimal Funky Jazz
Source images obtained from: https://www.zalando.fr/
Tip
For comprehensive video generation guidance, see the Video Generation guide.
Integration
P-Video integrates seamlessly with Pruna’s API infrastructure. P-Video requires an input image to work properly, so you’ll typically start by generating a seed image using P-Image, then use that image as input for video generation.
Tip
For more information on how to use the API, see the API Reference.
- API Endpoint
Base URL:
https://api.pruna.ai/v1/predictions
Authentication
Include your API key in the request header:
-H 'apikey: YOUR_API_KEY'
Step 1: Generate Input Image with P-Image
First, generate a seed image using P-Image that will serve as the input for video generation:
curl -X POST 'https://api.pruna.ai/v1/predictions' \
-H 'Content-Type: application/json' \
-H 'apikey: YOUR_API_KEY' \
-H 'Model: p-image' \
-d '{
"input": {
"prompt": "A person standing in a futuristic city at sunset, photorealistic",
"aspect_ratio": "16:9"
}
}'
Response includes a status URL to check progress:
{
"id": "1zww7deyssrme0csqwr90phzzr",
"model": "p-image",
"input": { },
"get_url": "https://api.pruna.ai/v1/predictions/status/1zww7deyssrme0csqwr90phzzr"
}
Check status and download the generated image:
curl -H 'apikey: YOUR_API_KEY' \
https://api.pruna.ai/v1/predictions/status/1zww7deyssrme0csqwr90phzzr
When completed, download the image:
curl -H 'apikey: YOUR_API_KEY' \
https://api.pruna.ai/v1/predictions/delivery/xezq/abc123.../output.jpg \
-o seed-image.jpg
Step 2: Upload Image for Video Generation
Upload the generated image (or any reference image) to use as input for P-Video:
curl -X POST https://api.pruna.ai/v1/files \
-H 'apikey: YOUR_API_KEY' \
-F 'content=@seed-image.jpg'
Response:
{
"id": "file-abc123",
"urls": {
"get": "https://api.pruna.ai/v1/files/file-abc123"
}
}
Use the file URL from the response in your video generation requests.
Step 3: Generate Video with P-Video
Asynchronous (Default)
Submit requests asynchronously for video generation using the uploaded image:
curl -X POST 'https://api.pruna.ai/v1/predictions' \
-H 'Content-Type: application/json' \
-H 'apikey: YOUR_API_KEY' \
-H 'Model: p-video' \
-d '{
"input": {
"prompt": "A person walking through a futuristic city at sunset, camera tracking from behind",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"duration": 10,
"aspect_ratio": "16:9",
"resolution": "720p",
"fps": 24
}
}'
Response includes a status URL to check progress:
{
"id": "1zww7deyssrme0csqwr90phzzr",
"model": "p-video",
"input": { },
"get_url": "https://api.pruna.ai/v1/predictions/status/1zww7deyssrme0csqwr90phzzr"
}
Check status by polling the status endpoint:
curl -H 'apikey: YOUR_API_KEY' \
https://api.pruna.ai/v1/predictions/status/1zww7deyssrme0csqwr90phzzr
When completed, download the result:
curl -H 'apikey: YOUR_API_KEY' \
https://api.pruna.ai/v1/predictions/delivery/xezq/abc123.../output.mp4 \
-o result.mp4
Synchronous (Try-Sync)
For faster results, use synchronous mode with the Try-Sync: true header:
curl -X POST 'https://api.pruna.ai/v1/predictions' \
-H 'Content-Type: application/json' \
-H 'apikey: YOUR_API_KEY' \
-H 'Model: p-video' \
-H 'Try-Sync: true' \
-d '{
"input": {
"prompt": "A person walking through a futuristic city at sunset, camera tracking from behind",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"duration": 5,
"draft": true
}
}'
If the generation completes within 60 seconds, you’ll receive the result directly:
{
"status": "succeeded",
"generation_url": "https://api.pruna.ai/v1/predictions/delivery/xezq/abc123.../output.mp4"
}
Download your video:
curl -H 'apikey: YOUR_API_KEY' \
https://api.pruna.ai/v1/predictions/delivery/xezq/abc123.../output.mp4 \
-o my-video.mp4
P-Image-(Edit) + P-Video
P-Video supports an efficient workflow for iterating on concepts and scaling up to production-quality videos. Note: P-Video requires an input image, so start by generating a seed image with P-Image.
Step 1: Generate Seed Image with P-Image
First, generate a seed image that will serve as the input for video generation:
curl -X POST 'https://api.pruna.ai/v1/predictions' \
-H 'Content-Type: application/json' \
-H 'apikey: YOUR_API_KEY' \
-H 'Model: p-image' \
-d '{
"input": {
"prompt": "A product on a white background, professional product photography",
"aspect_ratio": "16:9",
"seed": 42
}
}'
After downloading the generated image, upload it and use the file URL as image in your video generation requests.
Step 2: Draft Mode Iteration
Start with quick, low-quality iterations to test concepts and refine prompts:
{
"input": {
"prompt": "A product rotating on a white background",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"duration": 5,
"draft": true,
"seed": 42
}
}
This generates a draft video quickly, allowing you to iterate on prompts and concepts before committing to higher-quality generations.
Step 3: Scale Up to Production
Once satisfied with the concept, scale up to high-quality production videos with a fixed seed:
{
"input": {
"prompt": "A product rotating on a white background",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"duration": 10,
"resolution": "1080p",
"fps": 48,
"seed": 42
}
}
By using the same seed from your draft iteration, you maintain consistency while upgrading to production quality.
Custom Audio Support
Generate videos with custom audio:
{
"input": {
"prompt": "A person walking through a futuristic city at sunset",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"audio": "https://api.pruna.ai/v1/files/audio-abc123",
"save_audio": true
}
}
When audio is provided, the duration parameter is ignored and the video length matches the audio duration.
Configuration
P-Video offers flexible configuration options to customize video generation:
Tip
For more information on how to use the API, see the API Reference.
Required Parameters
Parameter |
Type |
Description |
|---|---|---|
prompt |
string |
Text prompt for video generation. |
image |
file |
Input image to generate video from (image-to-video). Supports jpg, jpeg, png, webp. |
Optional Parameters
Parameter |
Type |
Default |
Description |
|---|---|---|---|
audio |
file |
Input audio to condition video generation. Supports flac, mp3, wav. |
|
duration |
integer |
5 |
Duration of the video in seconds (1-10). Ignored when audio is provided. |
aspect_ratio |
string |
“16:9” |
Aspect ratio of the video. Ignored when an input image is provided. Allowed values: 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 1:1 |
resolution |
string |
“720p” |
Resolution of the video. Allowed values: 720p, 1080p |
fps |
integer |
24 |
Frames per second of the video. Allowed values: 24, 48 |
draft |
boolean |
false |
Draft mode. Generates a lower-quality preview of the video. |
prompt_upsampling |
boolean |
false |
Use prompt upsampling to enhance the prompt. |
save_audio |
boolean |
false |
Save the video with audio. |
seed |
integer |
Random seed. Set for reproducible generation. |
Example Configuration
Basic video generation:
{
"input": {
"prompt": "A product rotating on a white background",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"duration": 5,
"aspect_ratio": "16:9",
"resolution": "720p",
"fps": 24
}
}
Draft mode for rapid iteration:
{
"input": {
"prompt": "A product rotating on a white background",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"duration": 5,
"draft": true,
"seed": 42
}
}
With custom audio:
{
"input": {
"prompt": "A person walking through a futuristic city at sunset",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"audio": "https://api.pruna.ai/v1/files/audio-abc123",
"save_audio": true,
"aspect_ratio": "16:9"
}
}
High-quality 1080p at 48 FPS with prompt upsampling:
{
"input": {
"prompt": "A person walking through a futuristic city at sunset, camera tracking from behind",
"image": "https://api.pruna.ai/v1/files/file-abc123",
"duration": 10,
"resolution": "1080p",
"fps": 48,
"aspect_ratio": "16:9",
"prompt_upsampling": true,
"seed": 42
}
}