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

Pruna AI: P-Video-2

P-Video-2 is Pruna AI's quality-focused successor to P-Video, creating clips from text, images, or audio-conditioned inputs through one model. It produces sharper close-ups and foreground subjects, stronger identity and input-image consistency, improved lip synchronization for natively generated speech, and clearer on-screen text while generating dialogue, music, and sound effects with the video. It supports optional first-and-last-frame guidance, draft iteration, durations up to 20 seconds, 720p or 1080p output, and 24 or 48 FPS.
Compare models
p-video-2
AvailablePruna AIVideoAsync
Price
From $0.025
Modalities
Video

About P-Video-2

P-Video-2 is Pruna AI's quality-focused successor to P-Video. One model covers text-to-video, image-to-video, and audio-conditioned generation, with optional last-frame guidance. Pruna cites sharper close-ups, stronger identity consistency, better lip synchronization for generated speech, and clearer on-screen text. A draft mode allows cheaper previews, and output reaches 1080p at 24 or 48 frames per second.

Where it works well

  • Audio files can condition the generation, and the clip length then follows the audio.
  • Lip synchronization for natively generated speech is among the listed improvements over P-Video.
  • Draft mode renders low-cost previews so you can iterate on a prompt before the final render.
  • Frame rate can be set to 24 or 48, and resolution to 720p or 1080p.

When to choose another model

  • The quality tier takes longer than the original P-Video, which stays the pick for speed-first work.
  • It does not take video input; editing or replacing characters in footage is handled by separate Pruna models.

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.

    Audio to videoImage to videoStart and end frames to videoText 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 '{
      "duration": 5,
      "operation": "text-to-video",
      "model": "p-video-2",
      "prompt": "Cinematic sunrise over a calm lake with gentle camera motion."
    }'

Parameters

Audio to video

ParameterValues and limitsDefaultNotes
audio_urlNot providedNot providedAudio that drives the video: a public audio URL, 2 to 20 seconds. The output runs as long as the audio. Send operation: audio-to-video when the request also has an image; otherwise the request is read as image-to-video.
fps24, 4824Use an integer.
imageNot providedNot providedOptional first frame: a public image URL or a data URL. image, image_url and start_image are aliases; send one.
image_urlNot providedNot providedOptional first frame: a public image URL or a data URL. image, image_url and start_image are aliases; send one.
n11Use an integer.
operationNot providedNot providedNot provided
output_formatmov, mp4, webmmp4Not provided
promptNot providedNot providedMinimum length: 1 characters. Maximum length: 2048 characters.
resolution720p720pOnly 720p is offered.
seed≥ 0, ≤ 2147483647Not providedUse an integer.
start_imageNot providedNot providedOptional first frame: a public image URL or a data URL. image, image_url and start_image are aliases; send one.

Image to video

ParameterValues and limitsDefaultNotes
duration≥ 1, ≤ 20Not providedUse an integer.
fps24, 4824Use an integer.
imageNot providedNot providedFirst frame: a public image URL or a data URL. image, image_url and start_image are aliases; send one.
image_urlNot providedNot providedFirst frame: a public image URL or a data URL. image, image_url and start_image are aliases; send one.
n11Use an integer.
operationNot providedNot providedNot provided
output_formatmov, mp4, webmmp4Not provided
promptNot providedNot providedMinimum length: 1 characters. Maximum length: 2048 characters.
resolution720p720pOnly 720p is offered.
seconds≥ 1, ≤ 20Not providedUse an integer.
seed≥ 0, ≤ 2147483647Not providedUse an integer.
start_imageNot providedNot providedFirst frame: a public image URL or a data URL. image, image_url and start_image are aliases; send one.

Start and end frames to video

ParameterValues and limitsDefaultNotes
duration≥ 1, ≤ 20Not providedUse an integer.
end_imageNot providedNot providedLast frame: a public image URL or a data URL. Required.
fps24, 4824Use an integer.
n11Use an integer.
operationNot providedNot providedNot provided
output_formatmov, mp4, webmmp4Not provided
promptNot providedNot providedMinimum length: 1 characters. Maximum length: 2048 characters.
resolution720p720pOnly 720p is offered.
seconds≥ 1, ≤ 20Not providedUse an integer.
seed≥ 0, ≤ 2147483647Not providedUse an integer.
start_imageNot providedNot providedFirst frame: a public image URL or a data URL. Required.

Text to video

ParameterValues and limitsDefaultNotes
audio_urlInputs: ≤ 1Not providedOptional single source-audio URL. The upstream audio-input branch takes its duration from the audio and forbids an explicit duration; do not combine audio_url with duration or seconds. No audio-length bound is asserted here.
duration≥ 1, ≤ 20Not providedUse an integer.
fps24, 4824Use an integer.
height704, 768, 832, 960, 1088, 1152, 1280704Use an integer. Use a listed width and height pair: 1280x704 (720p, 16:9), 704x1280 (720p, 9:16), 1088x832 (720p, 4:3), 832x1088 (720p, 3:4), 1152x768 (720p, 3:2), 768x1152 (720p, 2:3), 960x960 (720p, 1:1).
n11Use an integer.
operationNot providedNot providedNot provided
output_formatmov, mp4, webmmp4Not provided
promptNot providedNot providedMinimum length: 1 characters. Maximum length: 2048 characters.
seconds≥ 1, ≤ 20Not providedUse an integer.
seed≥ 0, ≤ 2147483647Not providedUse an integer.
width704, 768, 832, 960, 1088, 1152, 12801280Use an integer. Use a listed width and height pair: 1280x704 (720p, 16:9), 704x1280 (720p, 9:16), 1088x832 (720p, 4:3), 832x1088 (720p, 3:4), 1152x768 (720p, 3:2), 768x1152 (720p, 2:3), 960x960 (720p, 1:1).

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.

Text to video · 720p

per second
Official price
$0.025
Verified price
—
Discount
—

Text to video · 1080p

per second
Official price
$0.05
Verified price
—
Discount
—

Image to video · 720p

per second
Official price
$0.025
Verified price
—
Discount
—

Image to video · 1080p

per second
Official price
$0.05
Verified price
—
Discount
—

Start and end frames to video · 720p

per second
Official price
$0.025
Verified price
—
Discount
—

Start and end frames to video · 1080p

per second
Official price
$0.05
Verified price
—
Discount
—

Audio to video · 720p

per second
Official price
$0.025
Verified price
—
Discount
—

Audio to video · 1080p

per second
Official price
$0.05
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-2 in Console with a prompt ready to edit or send.

Help me create with p-video-2 using text-to-video at /v1/videos/generations. Show the result, status, and cost.

Use cases

Best for
  • Text to Video
  • Talking scenes

    Generate a short scene with dialogue where the lips follow the generated speech.

  • Music-driven clips

    Give an audio track and let the picture follow its timing. The picture follows the audio timing.

  • Image animation

    Animate a brand image into a clip, optionally landing on a chosen last frame.

Prompt examples

Close-up of a barista pouring latte art, steam curling, she says 'Almost ready' with a smile. Café ambience.

Image-to-video: the attached mountain photo, clouds moving fast, camera slowly rising.

Use this voiceover audio. A founder walks through a sunlit office, gesturing as she speaks.

FAQ

What exactly is P-Video-2?

It is Pruna AI's premium video generation model, the quality-focused successor to P-Video. It creates clips from text, images, or audio inputs and generates dialogue, music, and sound effects with the picture.

How long can a clip be?

Pruna documents lengths up to twenty seconds, and the duration can be left out so the model chooses from the prompt. When you supply audio, the clip length follows the audio instead.

What resolutions and frame rates?

720p or 1080p at 24 or 48 frames per second. Aspect ratios such as 16:9, 9:16, 1:1 and 4:3 are available. Pick the frame rate that suits your delivery.

What is draft mode?

A faster, lower-quality preview setting for testing prompts. Use it to settle composition and motion, then rerun without it for the final clip. Draft mode is for testing, not delivery.

Which is better, P-Video-2 or P-Video-2-Pro?

P-Video-2 accepts audio input and offers 1080p, 48 fps and longer clips. P-Video-2-Pro is built on MiniMax H3, caps at 768p, and has no audio input.

How much does P-Video-2 cost?

On TokenLab, P-Video-2 costs $0.025-$0.05 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-2 use?

Use https://api.tokenlab.sh/v1/videos/generations for P-Video-2. The request example below shows the matching code shape.

Which operations does P-Video-2 support?

P-Video-2 supports Audio to video, Image to video, Start and end frames to video, Text to video. Select an operation above to see its endpoint and request example.

Compare P-Video-2

Guides that use P-Video-2

Sources

Reviewed Oct 2, 2026

More from Pruna AI

Related models