FrameSignal
Foundation video model

P-Video-2-Pro

Pruna's P-Video-2-Pro is an H3-based text/image-to-video model with generated audio, first-frame and first+last-frame conditioning, 480P/768P output, 5–15 second clips and cost/speed/quality modes. It is not a simple higher-spec P-Video-2: Pro trades P-Video-2's 1080P/20-second/48fps/audio-input surface for a different conditioning and quality-speed stack.

Official site View pricing Updated Sep 22, 2026
Tested by FrameSignal LabReviewed by FrameSignal EditorialTest date Testing pendingMethodology
Key facts

Verified model snapshot

ReleasedSep 17, 2026Verified Sep 22, 2026
Maximum duration5–15 secondsVerified Sep 22, 2026
Maximum resolution768PVerified Sep 22, 2026
Native audioYesVerified Sep 22, 2026
API statusavailableVerified Sep 22, 2026
Version delta

What changed

No structured version event has been recorded beyond the current public model entry.
Pricing & providers

Where to access this model

Provider

Pruna

12 observed offers

Last checked Sep 22, 2026

View 12 offer details
Pruna API · cost · 480P

USD 0.01 / returned video second

$0.01 per returned-video second

List price; explicit 50% promotion runs through Sep 24, 2026.

Source ↗
Pruna API · cost · 480P

USD 0.005 / returned video second

$0.005 per returned-video second during 50% promotion

Derived only from explicit official 50% off statement; re-verify after expiry.

Source ↗
Pruna API · speed · 480P

USD 0.02 / returned video second

$0.02 per returned-video second

List price; returned-duration billing.

Source ↗
Pruna API · speed · 480P

USD 0.01 / returned video second

$0.01 per returned-video second during 50% promotion

Derived from explicit 50% off statement.

Source ↗
Pruna API · quality · 480P

USD 0.04 / returned video second

$0.04 per returned-video second

List price; returned-duration billing.

Source ↗
Pruna API · quality · 480P

USD 0.02 / returned video second

$0.02 per returned-video second during 50% promotion

Derived from explicit 50% off statement.

Source ↗
Pruna API · cost · 768P

USD 0.025 / returned video second

$0.025 per returned-video second

List price; returned-duration billing.

Source ↗
Pruna API · cost · 768P

USD 0.0125 / returned video second

$0.0125 per returned-video second during 50% promotion

Derived from explicit 50% off statement.

Source ↗
Pruna API · speed · 768P

USD 0.035 / returned video second

$0.035 per returned-video second

List price; returned-duration billing.

Source ↗
Pruna API · speed · 768P

USD 0.0175 / returned video second

$0.0175 per returned-video second during 50% promotion

Derived from explicit 50% off statement.

Source ↗
Pruna API · quality · 768P

USD 0.075 / returned video second

$0.075 per returned-video second

List price; returned-duration billing.

Source ↗
Pruna API · quality · 768P

USD 0.0375 / returned video second

$0.0375 per returned-video second during 50% promotion

Derived from explicit 50% off statement.

Source ↗
Provider

WaveSpeedAI

4 observed offers

Last checked Sep 22, 2026

View 4 offer details
WaveSpeed API · speed · 480P

USD 0.02 / requested second

$0.02 per requested second

WaveSpeed bills requested duration, unlike Pruna returned-duration billing.

Source ↗
WaveSpeed API · quality · 480P

USD 0.04 / requested second

$0.04 per requested second

Same nominal rate does not imply same billing basis as Pruna.

Source ↗
WaveSpeed API · speed · 768P

USD 0.035 / requested second

$0.035 per requested second

WaveSpeed requested-duration billing.

Source ↗
WaveSpeed API · quality · 768P

USD 0.075 / requested second

$0.075 per requested second

WaveSpeed requested-duration billing.

Source ↗
Provider

EachLabs

4 observed offers

Last checked Sep 22, 2026

View 4 offer details
speed · 480P

USD 0.02 / provider generated second

$0.02 per generated second

Provider table also reports about 20s p50 runtime; keep semantics provider-specific.

Source ↗
quality · 480P

USD 0.04 / provider generated second

$0.04 per generated second

Provider table; p50 runtime around 20s.

Source ↗
speed · 768P

USD 0.035 / provider generated second

$0.035 per generated second

Provider table; p50 runtime around 20s.

Source ↗
quality · 768P

USD 0.075 / provider generated second

$0.075 per generated second

Provider table; p50 runtime around 20s.

Source ↗
Provider

Runware

5 observed offers

Last checked Sep 22, 2026

View 5 offer details
Runware videoInference

USD 0.01 / provider headline per second

from $0.01/s during launch promotion

Exact tier mapping is not inferred from the headline.

Source ↗
Runware videoInference · quality · 1344x768

USD 0.3797 / per example generation

$0.3797 for one 10-second example

Bronze Foundry Art-Pop example; about 1m9s end-to-end.

Source ↗
Runware videoInference · quality · 1344x768

USD 0.3797 / per example generation

$0.3797 for one 10-second example

Self-Rising Flood Barrier example; about 1m23s end-to-end.

Source ↗
Runware videoInference · speed · 768P

USD 0.0904 / per example generation

$0.0904 for one 5-second example

Tidal Accordion first-frame example; about 18s end-to-end.

Source ↗
Runware videoInference · quality · 768P

USD 0.5391 / per example generation

$0.5391 for one 15-second example

Cryogenic Seed Vault first+last-frame example; about 1m3s end-to-end.

Source ↗
Provider

Replicate

1 observed offer

Last checked Sep 22, 2026

View 1 offer details
Replicate predictions API · speed / quality · 480P / 768P

Official model updated in scan window; numeric price not captured.

Source ↗
Provider

Scenario

1 observed offer

Last checked Sep 22, 2026

View 1 offer details
speed / quality · 480P / 768P

Availability verified; raw price not captured.

Source ↗
Provider

Wiro

1 observed offer

Last checked Sep 22, 2026

View 1 offer details
POST https://api.wiro.ai/v1/Run/pruna/p-video-2-pro · 480P / 768P

Availability and API verified; raw price not captured.

Source ↗
Provider

Segmind

1 observed offer

Last checked Sep 22, 2026

View 1 offer details
POST /v2/p-video-2-pro · 480P / 768P

Availability and example runtime verified; raw price not exposed.

Source ↗
Provider

NanoGPT

1 observed offer

Last checked Sep 22, 2026

View 1 offer details
Provider offer

USD 0.1 / per generation

starting from $0.100 per video

Do not compare directly with per-second rates without task-specific normalization.

Source ↗
View detailed pricing
Derived signals

What stands out

Evidence-backed signals

  • Native audio is positively verified from an entity-supporting source.
  • Maximum duration is recorded as 5–15 seconds.
  • Official API access is verified as available.

Objective collection placement

Public test evidence

Recorded test runs

Hands-on testing pending. No quality score or first-hand verdict is shown until a complete raw test record is attached.
External evidence

Independent reports and examples

These observations belong to the named publishers and providers; they are not FrameSignal hands-on tests.

Runware · Runware

raw generation charge

Runware publishes a 10-second 1344x768 quality-mode bronze-foundry example with a raw $0.3797 charge, output URL and about 1m9s end-to-end time.

Reported metric: 0.3797 USD

Reported result: Provider displays about 1m9s end-to-end and exposes the generated output URL plus response cost.

Prompt: Bronze foundry art-pop music-video prompt; full prompt preserved at provider source.

Settings: duration seconds: 10 · resolution: 1344x768 · mode: quality · prompt upsampling: max · seed: 1061588653

Open source ↗
Runware · Runware

raw generation charge

Runware publishes a second 10-second 1344x768 quality-mode self-rising flood-barrier example with a raw $0.3797 charge and about 1m23s end-to-end time.

Reported metric: 0.3797 USD

Reported result: Provider displays about 1m23s end-to-end and publishes raw output.

Prompt: Self-rising flood-barrier demo prompt; full prompt preserved at provider source.

Settings: duration seconds: 10 · resolution: 1344x768 · mode: quality · prompt upsampling: max · seed: 1984884212

Open source ↗
Runware · Runware

raw generation charge

Runware publishes a 5-second 768P speed-mode first-frame example with a raw $0.0904 charge and about 18 seconds end-to-end time.

Reported metric: 0.0904 USD

Reported result: Provider displays about 18 seconds end-to-end and raw output.

Prompt: Tidal Accordion music-video prompt with supplied first frame; full prompt/input available at source.

Settings: duration seconds: 5 · resolution: 768P · mode: speed · seed: 1900679139

Open source ↗
Runware · Runware

raw generation charge

Runware publishes a 15-second 768P quality-mode first+last-frame example with a raw $0.5391 charge and about 1m3s end-to-end time.

Reported metric: 0.5391 USD

Reported result: Provider displays about 1m3s end-to-end and raw output.

Prompt: Cryogenic seed-vault B-roll prompt with first and last frames; full materials available at source.

Settings: duration seconds: 15 · resolution: 768P · mode: quality · prompt upsampling: turbo · seed: 1498766104

Open source ↗
Replicate · Pruna/Replicate official model

total job duration

Replicate exposes prompt, settings, output and a 10.8-second total job duration for an official P-Video-2-Pro example.

Reported metric: 10.8 seconds

Reported result: Replicate exposes prompt/settings/output and total job duration; exact price was not captured.

Prompt: Example prompt and settings preserved on Replicate's official Pruna model page.

Settings: provider: Replicate · model: prunaai/p-video-2-pro

Open source ↗
EachLabs · EachLabs

provider p50 runtime

EachLabs documents an 8-second 768P quality example and a provider p50 runtime around 20 seconds.

Reported metric: 20 seconds

Reported result: Provider documents a p50 runtime around 20 seconds and an example workflow.

Prompt: 8-second 768P quality product-film example; full prompt/settings available at provider page.

Settings: duration seconds: 8 · resolution: 768P · mode: quality

Open source ↗
Segmind · Segmind

generation time

Segmind documents a 6-second 768P speed example with approximately 49.98 seconds reported generation time.

Reported metric: 49.98 seconds

Reported result: Provider reports approximately 49.98 seconds for the documented example.

Prompt: Blacksmith-generation example; exact prompt/settings preserved at provider page.

Settings: duration seconds: 6 · resolution: 768P · mode: speed

Open source ↗
Capabilities

Capability matrix

Text to videoYes
Image to videoYes
Native audioYes
Reference imagesYes
Aspect ratios21:9, 16:9, 4:3, 1:1, 3:4, 9:16, adaptive
Compare with

Related public models

Grok Imagine Video 1.5

1–15 seconds canonical xAI generation; longer Roko buckets are provider-specific and non-official · API available

Open compare tool

H3 Max

5–15 seconds · API available

Open compare tool

Q3 Turbo

1–16s for T2V/I2V/start-end; 3–16s for reference-to-video · API available

Open compare tool
Alternatives

Related models to evaluate

Grok Imagine Video 1.5

Direct alternative: both are foundation model products with shared text to video, image to video, cinematic, native audio, character consistency workflows.

Open model

H3 Max

Direct alternative: both are foundation model products with shared text to video, image to video, cinematic, native audio, character consistency workflows.

Open model

Q3 Turbo

Direct alternative: both are foundation model products with shared text to video, image to video, cinematic, native audio, character consistency workflows.

Open model
Developer access

API

StatusavailableVerified Sep 22, 2026
PricingPruna lists 480P cost/speed/quality at $0.01/$0.02/$0.04 per returned-video second and 768P at $0.025/$0.035/$0.075; an explicit 50% promotion is active through 2026-09-24. WaveSpeed uses requested-duration billing, so nominal $/s is not directly equivalent.Verified Sep 22, 2026
Official accessDocumentation ↗
Open API record
Latest updates

P-Video-2-Pro change feed

View all updates
Evidence

Sources and verification

P-Video-2-Pro documentation
official docs · Official model source · Accessed Sep 22, 2026Pruna Developer Portal — P-Video-2-Pro
official docs · Official model source · Accessed Sep 22, 2026Introducing P-Video-2-Pro
official product · Official model source · Accessed Sep 22, 2026Pruna launches P-Video-2-Pro
official blog · Official model source · Accessed Sep 22, 2026Replicate — prunaai/p-video-2-pro
third party · Background only · Accessed Sep 22, 2026Replicate P-Video-2-Pro example
third party · Background only · Accessed Sep 22, 2026WaveSpeedAI P-Video-2-Pro
third party · Background only · Accessed Sep 22, 2026Runware P-Video-2-Pro
third party · Background only · Accessed Sep 22, 2026Runware P-Video-2-Pro examples
third party · Background only · Accessed Sep 22, 2026EachLabs P-Video-2-Pro
third party · Background only · Accessed Sep 22, 2026Scenario P-Video-2-Pro
third party · Background only · Accessed Sep 22, 2026Wiro P-Video-2-Pro API
third party · Background only · Accessed Sep 22, 2026Segmind P-Video-2-Pro
third party · Background only · Accessed Sep 22, 2026NanoGPT P-Video-2-Pro
third party · Background only · Accessed Sep 22, 2026Runware changelog — P-Video-2 Pro
third party · Background only · Accessed Sep 22, 2026P-Video-2 documentation
official docs · Background only · Accessed Sep 22, 2026