Choose Auto, TokenLab Verified, or Official for each request, with prices shown up front. See what's new

Pruna AI: P-Video-Avatar

P-Video-Avatar is a portrait-driven avatar video model that turns a single image into a speaking video using either an uploaded audio track or a generated voice from script. It is built for production avatar workflows with strong lip sync, selectable voices and languages, optional speaking-style control, seeded generation, and 720p or 1080p output for scalable talking-head video creation.
Compare models
p-video-avatar
AvailablePruna AIVideoAsync
Price
From $0.025
Modalities
Video

About P-Video-Avatar

P-Video-Avatar is Pruna AI's talking-head model: it turns one portrait image into a speaking video. You supply either an audio recording or a script that the model speaks with a chosen voice. Voices and languages are selectable, and optional prompts steer delivery and body movement. The portrait sets the first frame and the aspect ratio.

Where it works well

  • A single portrait is enough; no footage of the person is needed.
  • You can upload recorded audio for exact timing and pronunciation, or send a script for generated speech.
  • A voice prompt controls tone, pace and emotion, while a video prompt directs gestures and framing.
  • Pruna documents multiple voices and ten languages, plus a seed for reproducible variants.

When to choose another model

  • Very long clips can drift in consistency, so Pruna recommends keeping them short.
  • Lip sync is steadiest with a fixed camera; sweeping camera moves are a poor fit.

Getting started

  1. Create API key

    Create a key in Console, then use it with every model on the platform.

  2. Send your first request

    Copy the example for your language and run it against the endpoint.

    Image to videoTokenLab endpoint
    POST/v1/videos/generations
    curl -X POST "https://api.tokenlab.sh/v1/videos/generations" \
      -H "Authorization: Bearer sk-xxx" \
      -H "Content-Type: application/json" \
      -d '{
      "operation": "image-to-video",
      "model": "p-video-avatar",
      "prompt": "Cinematic sunrise over a calm lake with gentle camera motion.",
      "image_url": "https://example.com/image.jpg"
    }'

Parameters

Image to video

ParameterValues and limitsDefaultNotes
audio_urlInputs: ≥ 1, ≤ 1Not providedSupply one source-audio URL together with the portrait image. The active route requires inputs.audio and one frame image. The upstream script-to-speech alternative is outside this public parameter subset.
imageNot providedNot providedFirst frame (image, image_url or start_image): a public image URL or a data URL. Send one image.
image_urlNot providedNot providedFirst frame (image, image_url or start_image): a public image URL or a data URL. Send one image.
n11Use an integer.
operationNot providedNot providedNot provided
output_formatmov, mp4, webmmp4Not provided
promptNot providedNot providedNot provided
resolution720p, 1080p720pNot provided
seed≥ 0, ≤ 2147483647Not providedUse an integer.
start_imageNot providedNot providedFirst frame (image, image_url or start_image): a public image URL or a data URL. Send one image.

Fields outside this contract are rejected.

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Image to video · 720p

per second
Official price
$0.025
Verified price
—
Discount
—

Image to video · 1080p

per second
Official price
$0.045
Verified price
—
Discount
—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours
Requests
Success rate
P95 latency
Total tokens
Model performance
Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open P-Video-Avatar in Console with a prompt ready to edit or send.

Help me create with p-video-avatar using image-to-video at /v1/videos/generations. Show the result, status, and cost.

Use cases

  • Spokesperson videos

    Turn a headshot and a script into a presenter clip for onboarding or product updates.

  • Localised messages

    Re-voice one portrait in several languages for regional versions. Replace only the script and voice to make each version.

  • Support explainers

    Create a friendly avatar that walks users through a help-centre article. Pair it with a short written guide or tutorial video.

Prompt examples

Script: 'Welcome to the team. Your first week starts Monday at nine.' Voice: warm, calm pace. Video: eye-level, small hand gestures.

Upload a 20-second recorded podcast intro; keep the host in a fixed shot, subtle head nods.

Script in Spanish: product announcement with an upbeat tone, bright office background.

FAQ

What exactly is P-Video-Avatar?

It is a Pruna AI model that creates a speaking avatar video from a single image and either uploaded audio or a text script. It focuses on lip sync for talking-head content.

Should I use audio or a script?

Both work. Use a script for scale and generated voices, or upload audio when exact pronunciation matters. If both are sent, Pruna says the uploaded audio takes priority.

Which voices and languages?

Pruna's guide lists 30 voices and ten languages, including English, Spanish, French, German, Italian, Portuguese, Japanese, Korean and Hindi. Match voice gender and language to the face.

Can I control the performance?

Yes. A voice prompt shapes tone and pace, and a video prompt guides movement, framing, and background behaviour. A seed lets you compare variants one change at a time.

What output sizes are offered?

720p or 1080p. Pruna suggests drafting at 720p and finishing at 1080p, and keeping clips short for consistent results. Iterate cheaply first, then render the final version once. Keep the same seed while comparing.

How much does P-Video-Avatar cost?

On TokenLab, P-Video-Avatar costs $0.025-$0.045 per second. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should P-Video-Avatar use?

Use https://api.tokenlab.sh/v1/videos/generations for P-Video-Avatar. The request example below shows the matching code shape.

Which operations does P-Video-Avatar support?

P-Video-Avatar supports Image to video. Select an operation above to see its endpoint and request example.

Compare P-Video-Avatar

Sources

Reviewed Oct 2, 2026

More from Pruna AI

Related models