Is P-Video 2 Pro uncensored for NSFW video?

Yes. There is no filter on prompts, images or results. In our tests the content prompts we ran rendered as written on all three endpoints; requests at the highest levels came back toned down, which is a limit of the model rather than a refusal.

In our tests a 5-second clip at 480p took about 10 to 20 seconds of generation time, and reference clips with a video took around 20 seconds. Queueing and delivery add a little on top.

What is the difference between cost, speed and quality?

They are price tiers set by Pruna. cost is the cheapest per second and a little slower; speed is faster at the same picture quality; quality renders a higher-quality clip at the highest price. Reference to video offers cost and speed only.

Does every clip have sound?

Yes. The model generates ambience, effects and any line you write in quotes. On reference to video you can use your own audio clip as the soundtrack instead with keep_reference_audio.

How do I refer to my uploads in the prompt?

On reference to video, write video 1, video 2 for reference videos and audio 1, audio 2 for audio clips, numbered in upload order. Opening and closing frames are not named; they are used as the first and last frame.

Per second of output; the rate depends on resolution and generation mode. Reference to video also bills the seconds of reference videos, a small charge per opening or closing frame and a flat charge for reference audio. Failed requests are not charged.

P-Video 2 Pro API

pruna-ai/p-video-2-pro/text-to-video

P-Video 2 Pro text-to-video turns a written brief into a 5-15 second clip with sound at 480p or 768p, in seven aspect ratios, in about 10 to 20 seconds of generation time. No content filter.

From $0.012/s

Run model View API

Input

Text to video

Prompt

Aspect ratio

Resolutionaffects price

Generation Modeaffects price

costspeedquality

Durationaffects price

Prompt optimization

offturbomax

Seed

Safety checkerOff

SpicyAPI does not run a safety checker on this model. Your prompt and media go to the model as written, and the model itself decides what comes back.

Checking your account before enabling submission and pricing.

Output

Ready for a new run

Example result

Extend

Video upscaling

Video editing

{
  "model": "pruna-ai/p-video-2-pro/text-to-video",
  "input": {
    "aspect_ratio": "16:9",
    "resolution": "768p",
    "generation_mode": "speed",
    "duration_seconds": 5,
    "prompt_optimization_mode": "off",
    "prompt": "Night, inside an art-deco cocktail bar carved into a sea cliff, brass lamps and green marble, black sea beyond tall arched windows. Slow dolly-in along the polished bar toward a bartender in her forties with a silver bob, wearing a black waistcoat and white shirt; she slides a martini across the marble to a man in a white dinner jacket seated at the end of the bar, raises one eyebrow and says, \"Same as last summer?\" He smiles and lifts the glass. Warm amber light, shallow depth of field, 35mm film look. Sound: a soft jazz piano in the corner, ice clinking, waves breaking far below."
  }
}

Introduction

P-Video 2 Pro Text to video turns a written brief into a clip of 5 to 15 seconds at 480p or 768p, in seven aspect ratios from 16:9 to 9:16. Pruna built P-Video 2 Pro for speed: in our tests a 5-second clip took about 10 to 20 seconds to generate. Every clip comes back with a soundtrack the model generates in the same pass, and there is no content filter on the prompt or the result.

Choose a P-Video 2 Pro mode by starting material

What you have at the start	Mode
a written scene or script	P-Video 2 Pro Text to video (you are here)
an opening frame, and optionally a closing frame	P-Video 2 Pro Image to video
a clip to borrow motion from, or a voice or music track	P-Video 2 Pro Reference to video

Key features & highlights

Three generation modes

generation_mode trades price against speed and quality. cost is the cheapest per second and a little slower, speed (the default) is faster at the same picture quality, and quality renders a higher-quality clip at a higher price.

Sound on every clip

The model writes the soundtrack itself: ambience, effects and any line you put in quotes. There is no switch and no extra charge for it.

Uncensored

No filter on prompts or results. In our tests prompts up to the Strong level rendered as written; requests at the highest levels came back toned down, which is a limit of the model, not a refusal.

Request parameters

Request parameters for P-Video 2 Pro · Text to Video
Parameter	Required	Description
`prompt`	Yes	Describe the subject, the action, the camera and the sound. Put spoken lines in quotes and the sound on its own line. at least 1 characters
`aspect_ratio`	No	16:9 (default), 9:16, 4:3, 3:4, 3:2, 2:3 or 1:1. It does not change the price. `16:9` · `9:16` · `4:3` · `3:4` · `3:2` · `2:3` · `1:1` · Default: `16:9`
`resolution`	No	480p or 768p (default 768p). 768p costs more per second. `480p` · `768p` · Default: `768p`
`generation_mode`	No	cost (cheapest, a little slower), speed (default, faster at the same quality) or quality (best picture, highest price). `cost` · `speed` · `quality` · Default: `speed`
`duration_seconds`	No	Whole seconds, 5 to 15 (default 5). Drives the price. `≥ 5` · `≤ 15` · Default: `5`
`prompt_optimization_mode`	No	off (default) sends your prompt word for word; turbo and max let the service rewrite it first. Does not change the price. `off` · `turbo` · `max` · Default: `off`
`seed`	No	Keeps runs comparable while you change the prompt. It does not promise a frame-identical clip. `≥ 0` · `≤ 2147483647`

Pricing

Billed per second of output. The price per second depends on the resolution (480p or 768p) and the generation mode: cost is the cheapest, speed costs more, quality the most. Prompt expansion, aspect ratio, seed and the closing frame do not change the price. A failed request is not charged.

Pricing
Resolution	Generation mode: cost	Generation mode: speed	Generation mode: quality
`480p`	$0.012 / second	$0.024 / second	$0.048 / second
`768p`	$0.03 / second	$0.042 / second	$0.09 / second

Use cases

Fast drafts

Clips come back in seconds, so you can try five versions of a shot in the time other models take for one. Draft at 480p in cost mode, then rerun the prompt you like at 768p.

Short scenes with a spoken line

Describe the action, then put the line in quotes and the ambience after Sound:. One clear line per clip lands best.

Vertical clips for social

Choose 9:16 or 3:4 and frame the subject in the middle of the shot; the model composes for the vertical frame.

Explore key use-case tutorials

Prompt guide & pro tips

Write it in the order it happens

Subject and wardrobe, action, camera move, light, then a separate Sound: line.

Keep prompt expansion off for exact wording

prompt_optimization_mode is off by default, so the prompt goes to the model word for word. turbo and max let the service rewrite it first, which can add detail but may also change what you asked for.

Pick the mode last

Find the framing in cost mode, then switch to speed for faster turnaround or quality for the final render.

API quick start

# Generate the idempotency key once; reuse it when retrying after a lost response.
export SPICY_IDEMPOTENCY_KEY="${SPICY_IDEMPOTENCY_KEY:-$(uuidgen)}"
curl -X POST https://api.spicyapi.ai/api/v1/jobs/createTask \
  -H "Authorization: Bearer $SPICY_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $SPICY_IDEMPOTENCY_KEY" \
  -d '{
  "model": "pruna-ai/p-video-2-pro/text-to-video",
  "input": {
    "prompt": "A slow aerial shot drifting over a misty pine forest at sunrise",
    "resolution": "768p",
    "aspect_ratio": "16:9",
    "generation_mode": "speed",
    "duration_seconds": 5,
    "prompt_optimization_mode": "off"
  }
}'

# Copy data.taskId from the response into TASK_ID, then poll the status.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.spicyapi.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $SPICY_API_KEY"

See the API reference for authentication, the complete schema, polling, and error handling.

Missing the model you need?

Tell us what to add and where you would use it. Requests go to the people who decide what gets onboarded next.

FAST VIDEO WITH SOUND

P-Video 2 Pro API at a glance

Pruna's fastest video model: 5 to 15 second clips with sound in seconds, from text, frames or reference clips, with no content filter.

Generation time: ~15 s; A 5-second clip took about 10 to 20 seconds to generate in our tests.
Clip length: 5-15 s; Any whole number of seconds, billed per second.
Modes: 3; cost, speed or quality: pick price, turnaround or picture quality.
On every clip: Sound; Ambience, effects and quoted lines, generated with the picture.
Content filter: None; Uncensored: no filter on prompts, images or results.

Results

What you can make with P-Video 2 Pro

From references
Borrow a camera move
The camera copies the slow push-in of video 1 while the pier, the launch and the woman all come from the opening image - nothing from the bar in the reference clip shows up. She walks forward from the stern, pulls the wrap around her shoulders and ends facing the camera. The soundtrack is very quiet: soft water against the hull, no music audible.
Reference to video: an opening image plus a 4-second reference clip named in the prompt as video 1, 5 seconds at 768p.
From text
A scene with its own sound
One continuous shot with a gentle drift toward the pair rather than a strong dolly. The line "Same as last summer?" is spoken clearly near the end, after a music bed in the first half; he picks up the glass and smiles on the last beat.
Text to video, 5 seconds at 768p, 16:9, with one spoken line and music generated on the soundtrack.
From frames
Pin the start and the end
The clip starts exactly on the opening frame and lands on the closing frame: he folds the letter, tucks it away, rises and moves to the balustrade while the fireworks start about a second and a half in. The fixed camera holds the same framing throughout. The soundtrack is crowd applause and crackle rather than distinct firework bangs.
Image to video with an opening and a closing frame, 5 seconds at 768p.

Content test

Is P-Video 2 Pro uncensored? What we tested

We ran each level on our live setup and checked every clip frame by frame. A pass means the content was actually on screen, not just that the request went through.

Content levels tested on P-Video 2 Pro, by mode
Level	Text to video	Image to video	Reference to video
MildThe lightest prompts in the test	Not tested yet	Not tested yet	Not tested yet
ModerateWhere filtered models start to refuse	RendersTested Oct 17, 2026	RendersTested Oct 17, 2026	RendersTested Oct 17, 2026
StrongThe middle of the scale	RendersTested Oct 17, 2026	Not tested yet	Not tested yet
Very strongWhere most models start to hold back	Not tested yet	Not tested yet	Not tested yet
MaximumThe top of the scale	Not tested yet	Not tested yet	Not tested yet

Renders: on screen as written. Softened: generated, but milder than what was written. Refused: rejected. Not tested yet: we have not run that level on that mode.

How each mode behaves

All modes: Clips are 480p or 768p and 5 to 15 seconds; there is no 1080p. Every clip carries generated audio unless you supply your own on reference to video.
All modes: In our tests prompts at the highest levels came back toned down rather than as written. That is a limit of the model; nothing is refused.
Image to video: A prompt is required, and the clip keeps the shape of the opening frame; there is no aspect-ratio choice.
Reference to video: Reference videos must be 2 seconds or longer, 15 seconds in total, and their seconds are billed. Reference audio needs an opening frame or a reference video too. Reference images other than the opening and closing frame are not available on this endpoint.

Use cases

What to build with the P-Video 2 Pro API

Rapid iteration
Clips come back in seconds, so you can try many takes of a shot and keep the best one.
Uncensored scenes
Nothing is filtered. In our tests prompts up to the Strong level rendered as written; requests at the highest levels came back toned down.
Restaging a reference
Borrow the camera move of a clip you like and put your own character in it with an opening frame.

Prompts

P-Video 2 Pro prompt examples you can copy

Paste one into the playground as it is, then change one thing at a time.

Reference to video
Name the clip you borrow from
Result from this prompt
Start from the opening image: the woman in the midnight-blue satin gown at the stern of the mahogany motor launch, moored at the lantern-lit pier below the glowing cliff hotel. Follow the camera move and pacing of video 1 - the same slow, steady dolly-in at the same unhurried speed - but keep the pier, the launch and the woman from the image; do not copy the bar from video 1. As the camera glides closer she turns her head toward us, draws the silver wrap closer around her shoulders against the sea breeze and gives a slow, knowing smile, the lanterns swaying slightly overhead. Sound: water lapping against the hull, lantern ropes creaking, faint music drifting down from the hotel.
Why it works. Naming video 1 tells the model to take the motion from the clip while the opening frame sets who is in the shot.
Try this prompt
Text to video
Action, camera, then sound
Result from this prompt
Night, inside an art-deco cocktail bar carved into a sea cliff, brass lamps and green marble, black sea beyond tall arched windows. Slow dolly-in along the polished bar toward a bartender in her forties with a silver bob, wearing a black waistcoat and white shirt; she slides a martini across the marble to a man in a white dinner jacket seated at the end of the bar, raises one eyebrow and says, "Same as last summer?" He smiles and lifts the glass. Warm amber light, shallow depth of field, 35mm film look. Sound: a soft jazz piano in the corner, ice clinking, waves breaking far below.
Why it works. The prompt runs in the order things happen and keeps the sound on its own line, so the action and the audio both land.
Try this prompt
Image to video
Pin the ending with a second frame
Result from this prompt
He lowers the letter, smiles to himself, folds it and tucks it into his breast pocket, then stands, steps around the marble table to the stone balustrade and rests both hands on it as golden fireworks begin to burst over the harbour across the bay. Locked-off camera, warm candlelight on his face, the reflections of the fireworks dancing on the black water. Sound: the first fireworks thudding and crackling across the bay, distant applause from the harbour, gentle waves.
Why it works. With a closing frame set, the prompt only needs to describe how the shot gets from one frame to the other.
This mode starts from your own image: add it in the playground before you run.
Try this prompt

Pricing

P-Video 2 Pro API pricing

A 5-second clip starts at $0.06 at 480p and $0.15 at 768p.

The playground shows the exact price for your settings before you run. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.

See every price tier

Get started

How to use the P-Video 2 Pro API in three steps

Step 1
Draft cheap, finish sharp
Find the shot at 480p in cost mode, then rerun the same prompt at 768p in speed or quality mode.
Step 2
Write a Sound: line
Every clip has audio. Name the ambience and put any spoken line in quotes so you get the sound you want.
Step 3
Name your references
On reference to video, write "video 1" or "audio 1" in the prompt and say what to take from each.

Open the playground Get an API key

Parameter reference and code samples are in the API and Readme tabs at the top of this page.

Why SpicyAPI

Why run P-Video 2 Pro on SpicyAPI

No filter from us
We add no content filter of our own. What the model renders is what you get back.
Tested, not guessed
Content labels come from real generations we ran and checked, with the test date on the page.
Pay per result, in dollars
No subscription and no credits. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.
One key for every model
One API key and balance cover image, video, audio and chat models, so you can explore more models from the same account.

FAQ

P-Video 2 Pro API FAQ

About P-Video 2 Pro

P-Video 2 Pro brings Pruna AI's fast video model to one API: text to video, image to video with an opening and closing frame, and reference to video from clips and audio. Clips run 5 to 15 seconds at 480p or 768p with a generated soundtrack, in three generation modes, with no content filter. SpicyAPI bills per second of output in US dollars, with no subscription.

Workbench

Input

Output

Introduction

Choose a P-Video 2 Pro mode by starting material

Key features & highlights

Three generation modes

Sound on every clip

Uncensored

Request parameters

Pricing

Use cases

Fast drafts

Short scenes with a spoken line

Vertical clips for social

Prompt guide & pro tips

Write it in the order it happens

Keep prompt expansion off for exact wording

Pick the mode last

API quick start

Borrow a camera move

A scene with its own sound

Pin the start and the end

How each mode behaves

Rapid iteration

Uncensored scenes

Restaging a reference

Name the clip you borrow from

Action, camera, then sound

Pin the ending with a second frame

Draft cheap, finish sharp

Write a Sound: line

Name your references

No filter from us

Tested, not guessed

Pay per result, in dollars

One key for every model

Is P-Video 2 Pro uncensored for NSFW video?+

How fast is it?+

What is the difference between cost, speed and quality?+

Does every clip have sound?+

How do I refer to my uploads in the prompt?+

How is it priced?+

About P-Video 2 Pro

Related models and next tasks