What can I create with P-Video 2 Pro Avatar?

P-Video 2 Pro Avatar can create narrated product demos, presenter clips and multilingual talking videos on each::labs. Provide a portrait with your script, then choose a voice and supported language. Delivery instructions guide tone and pacing without being spoken, while a visual prompt describes what the video should show.

How are P-Video 2 Pro Avatar videos priced?

Pruna prices P-Video 2 Pro Avatar by the duration of the generated video: $0.02 per second at 480p or $0.04 per second at 768p. Uploaded audio sets the video length and takes priority over a script. Voice and language do not change the rate, and the maximum video length is three minutes.

Example inputhover

image
voice: "Kore (Female)"
prompt: "One single continuous shot with the exact framing of the image, a static camera, no cuts, no zoom. The woman keeps looking directly into the camera for the entire video and talks to the viewer with clear lip movement on every word, warm confident smile, small natural head movements. She holds the laptop still in front of her waist and never looks down at it. The laptop lid stays completely plain pastel pink with no logo. Same outfit, flowers and lighting as the image. Photorealistic."
resolution: "768p"
voice_prompt: "Say the following in a warm, upbeat, confident presenter voice, friendly and natural. Give a proud smile on the last sentence."
voice_script: "Hi, I'm Mia. Pruna P Video 2 Pro Avatar just landed on Eachlabs. Upload a photo, write a script, and your presenter is ready to talk, in any language. Close the imagination gap with Eachlabs."
voice_language: "English (US)"
prompt_upsampling: true

Pruna | P-Video 2 Pro | Avatar

Video Gen·P-video·by Pruna AI

P-Video 2 Pro Avatar turns portraits and audio or scripts into talking videos with synced lips. Create narrated demos and multilingual presenter clips.

Try it now →

API reference

Runtime (p50): -
Estimated price: $0.02–$0.04 per second, depending on resolution

Call the API

prediction.sh

curl -X POST \
  -H "Authorization: Bearer $EACHLABS_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "p-video-2-pro-avatar",
    "input": {
        "image": "https://cdn-us.eachlabs.ai/defaults/876cff25cc7d47d5a60c5a2ed1a4a7db.png",
        "voice": "Kore (Female)",
        "prompt": "One single continuous shot with the exact framing of the image, a static camera, no cuts, no zoom. The woman keeps looking directly into the camera for the entire video and talks to the viewer with clear lip movement on every word, warm confident smile, small natural head movements. She holds the laptop still in front of her waist and never looks down at it. The laptop lid stays completely plain pastel pink with no logo. Same outfit, flowers and lighting as the image. Photorealistic.",
        "resolution": "768p",
        "voice_prompt": "Say the following in a warm, upbeat, confident presenter voice, friendly and natural. Give a proud smile on the last sentence.",
        "voice_script": "Hi, I'm Mia. Pruna P Video 2 Pro Avatar just landed on Eachlabs. Upload a photo, write a script, and your presenter is ready to talk, in any language. Close the imagination gap with Eachlabs.",
        "voice_language": "English (US)",
        "prompt_upsampling": true
    },
    "webhook_url": ""
}' \
  https://api.eachlabs.ai/v1/prediction/

Related models

4 models

Smash and Grab VideoUnified AI API for hundreds models. Access video, image, and audio generation from leading providers through one integration with eachlabs.

Gas Station DanceUnified AI API for hundreds models. Access video, image, and audio generation from leading providers through one integration with eachlabs.

FLUX 3 Max Image to Video AI model preview

FLUX 3 Max Image to VideoBlack Forest Labs

Pruna | P-Video 2 ProPruna AI

* FAQ

About Pruna | P-Video 2 Pro | Avatar

01 / 03

What is P-Video 2 Pro Avatar?

P-Video 2 Pro Avatar generates a talking video from a portrait and speech. You can provide an audio recording or a script for synthesized speech, with synchronized lips and expressions. Pruna returns a 24 fps MP4 lasting between one second and three minutes, with the portrait determining the aspect ratio.