P Video 2 on Pixel Dojo
P Video AI Video Generator
P Video 2 makes 1080p clips that arrive with their own sound: spoken lines, sound effects and music, all from one prompt.
P Video 2 is a fast text to video and image to video model that writes the soundtrack with the picture. Ask a character to say a line and she says it, word for word. Ask for sparks, rain or a jazz piano and you hear them. Clips run up to 20 seconds at 720p or 1080p, at 24 or 48 frames per second. We made every example on this page on Pixel Dojo, and each one comes with the exact prompt and settings behind it.
3.75 credits ($0.23) per 5 s clip at 720p. Included in every plan, from $10/month.
Already a member? Sign in
By Brian Lovett · Updated
P Video examples: action, products, aerials and animation
Six 1080p clips we made with P Video 2, each with its own native soundtrack. Unmute a tile to hear it, then tap Use this prompt to start from it.
Action at 1080p, 5 seconds. Every hammer strike throws sparks through the smoke, and you hear the ring of steel on the anvil.
Wide low angle shot, slow push in. A blacksmith in his forties with a short dark beard, a scarred leather apron and rolled sleeves strikes a glowing orange bar of steel on the anvil with a heavy hammer, each blow throwing a bright fountain of sparks across the dark forge, embers drifting through a shaft of smoky daylight. 35mm lens, high contrast, warm orange against deep teal shadows. Audio: loud ringing hammer strikes recorded close, the roar of the forge fire, crackling embers.
Aerial at 1080p, 10 seconds. A low, fast drone pass that rises to reveal the sea stacks, with the roar of the surf underneath.
Aerial drone shot gliding low and fast over a black sand beach as long white waves roll in, then rising to reveal jagged basalt sea stacks and a distant glacier under a low golden sun, mist drifting off the surf. Wide 24mm lens, rich contrast, cool blues and warm gold. Audio: crashing surf, strong wind, a swelling orchestral string bed.
Product shot at 1080p, 1:1, 5 seconds. Slow motion splashes and golden glass that stays perfectly steady, ready for a social ad.
Macro product shot, slow push in. A square amber glass perfume bottle with a polished gold cap stands on wet black slate as drops of water fall in slow motion and burst into glittering splashes around it, a single soft light sweeping across the glass. Deep black background, crisp reflections, 100mm macro lens. Audio: crisp droplet splashes recorded close, a soft airy synth pad.
3D animation at 1080p, 6 seconds. Soft global light, blooming flowers and a playful plucked string melody with chimes.
Stylized 3D animated film, slow orbit. A round little white robot gardener with big glowing blue eyes waters a bed of glowing pink and gold flowers on a tiny floating island at sunset, the flowers blooming open as the water touches them, waterfalls spilling off the island edge into the clouds below. Soft global illumination, vivid saturated color, shallow depth of field. Audio: trickling water, gentle chimes as each flower opens, a whimsical plucked string melody.
Dance at 1080p, 9:16, 48 fps. Silk that flows and catches the light, made vertical for Reels, TikTok and Shorts.
Vertical medium wide shot, slow orbit. A dancer in her late twenties with a short platinum bob spins on a city rooftop at golden hour in a flowing saffron yellow silk dress, the skirt flaring wide around her, the skyline glowing behind her, warm backlight through her hair. 35mm lens, shallow depth of field. Audio: the swish of silk, light footsteps on concrete, an upbeat house beat.
Night scene at 1080p, 5 seconds. The neon sign reads OPEN ALL NIGHT, with rain, a sizzling grill and a faint jukebox song.
Night exterior, slow dolly in. A rain soaked corner diner with a glowing red neon sign that reads "OPEN ALL NIGHT" buzzing above the big front window, a cook in a white paper hat flipping burgers on the grill inside, reflections shimmering in the puddles. 35mm anamorphic lens, teal and red grade. Audio: neon buzz, rain on pavement, sizzling grill, a faint doo wop song from a jukebox.
How to write a P Video prompt that sounds as good as it looks
- 1
Open with the shot
Start with shot size and camera move: "Medium close-up, slow push in" or "Aerial drone shot gliding low and fast". P Video holds that move for the whole clip.
- 2
Describe one subject and one action
Give the subject a few specific details and one clear beat: a blacksmith strikes a glowing bar, a robot waters a flower bed, a dancer spins.
- 3
Write the dialogue in quotes
For speech, write it like a script: She smiles and says, "This one goes out to everyone still awake." P Video speaks the line as written.
- 4
End with an Audio line
List the sounds you want: "Audio: neon buzz, rain on pavement, sizzling grill, a faint doo wop song from a jukebox." Each of our clips follows its Audio line.
- 5
Pick length, resolution and frame rate
1 to 20 seconds, 720p or 1080p, 24 or 48 fps. Use 48 fps for fast motion like dance and sports, and 9:16 for vertical social video.
- 6
Add a start image if you have one
Upload a first frame to animate a product photo or a character you already made, or add an end frame to steer where the shot lands.
P Video dialogue test: one line, spoken word for word
We wrote the radio host her line in quotes and asked for a warm close voice, vinyl crackle and a jazz piano. We ran it at 720p for 5 seconds and at 1080p for 8 seconds with a fader move at the end. A speech to text check of both soundtracks came back with the exact line: "This one goes out to everyone still awake."
1080p, 8 seconds, 12 credits. The line, a smile and a fader push as the piano swells.
Medium close-up, slow push in. A night shift radio host in her fifties with a silver bob and tortoiseshell glasses leans into a single vintage chrome microphone in a dim studio lit by a red ON AIR lamp and glowing amber meters. She smiles and says, "This one goes out to everyone still awake." Then she slides a fader up as the music swells. 85mm lens, shallow depth of field, warm film grain. Audio: her warm low voice recorded close, soft vinyl crackle, a quiet jazz piano that swells at the end.
720p, 5 seconds, 3.75 credits. The same line in a tighter cut.
Medium close-up, slow push in. A night shift radio host in her fifties with cropped silver hair and tortoiseshell glasses leans into a vintage chrome microphone in a dim studio lit by a red ON AIR lamp and glowing amber meters. She smiles and says, "This one goes out to everyone still awake." 85mm lens, shallow depth of field, warm film grain. Audio: her warm low voice recorded close, soft vinyl crackle, a quiet jazz piano under her.
- Put spoken words in double quotes after "says". Both clips speak the line exactly as written.
- Describe the voice as well as the words: "her warm low voice recorded close" gave a close, intimate read.
- Layer the room: vinyl crackle and a quiet piano under the voice make the clip feel finished without any editing.
- Use 1080p when the face fills the frame. Glasses, hair and skin hold their detail through the push in.
What to make with P Video
Talking characters and presenters
Write the line in quotes and P Video speaks it with matching mouth movement, for intros, explainers and character moments.
Product and social ads
Slow motion splashes, light sweeps and macro detail with sound design included, in 1:1, 9:16 or 16:9.
Aerial and travel shots
Drone passes and reveals up to 20 seconds long, with wind and surf already on the track.
Dance and music video moments
Spins, flowing fabric and rooftop light at 48 fps for smooth, crisp motion.
Stylized and 3D animation
Cute characters, glowing worlds and whimsical scores for shorts, intros and kids content.
Atmospheric scene setters
Rainy streets, neon signs and busy kitchens that set a mood in a few seconds, sound and all.
P Video settings at a glance
These are the controls in the P Video tool and the values behind our clips.
- ModelP Video 2 (default), P Video 1
- P Video 2 is the newest and what every clip on this page uses. P Video 1 is the lower cost previous generation.
- Duration1 to 20 seconds
- Our clips run 5, 6, 8 and 10 seconds. With an uploaded audio track the clip runs as long as the audio, up to 20 seconds.
- Resolution720p, 1080p
- Our 1080p files measure 1920 x 1088 at 16:9 and 1088 x 1920 vertical.
- Frame rate24 or 48 fps
- 24 fps for a film look, 48 fps for fast motion. Our dancer clip runs at 48 fps.
- Aspect ratio16:9, 9:16, 1:1 and more
- Pick the frame for the platform: 16:9 for YouTube, 9:16 for Reels and TikTok, 1:1 for feeds.
- Start and end imageOptional
- Animate a first frame you already have, or set the last frame the shot should land on.
- AudioGenerated, or upload your own
- P Video writes sound to match the prompt. Upload a voice or music track and the video follows it.
- DraftOn or off
- Draft mode renders a lower cost preview at 0.45 credits a second at 720p, handy for checking an idea.
P Video pricing: from 0.75 credits a second
P Video is included in every Pixel Dojo plan, from $10/month. P Video 2 costs 0.75 credits per second at 720p and 1.5 credits per second at 1080p, sound included.
That makes a 5 second 720p clip 3.75 credits and a 5 second 1080p clip 7.5 credits. Our 10 second 1080p aerial cost 15 credits.
We made every clip on this page through the Pixel Dojo API at those prices, and the same prompts cost the same in the web tool.
P Video questions
What is P Video?
P Video is a fast AI video model for text to video and image to video. P Video 2, the version on every clip on this page, renders up to 20 seconds at 720p or 1080p and generates the soundtrack with the picture, including dialogue, sound effects and music.
Does P Video make sound and dialogue?
Yes. Write spoken lines in quotes and list sounds in an Audio line. Our radio host speaks her line word for word, and our forge, surf, diner and robot clips each come with their own sound effects and music.
How long can a P Video clip be?
Up to 20 seconds. Set the duration from 1 to 20 seconds, or upload an audio track and the clip runs as long as the audio, up to 20 seconds.
What resolution and frame rate does P Video support?
720p and 1080p, at 24 or 48 fps. Our 1080p clips measure 1920 x 1088 in 16:9 and 1088 x 1920 vertical.
How much does P Video cost?
P Video is included in every plan, from $10/month. P Video 2 is 0.75 credits per second at 720p and 1.5 credits per second at 1080p, so a 5 second 1080p clip is 7.5 credits.
Can I use P Video through an API?
Yes. POST a prompt with duration, resolution and aspect_ratio to /api/v1/models/p-video/run, and add an image or audio URL to animate a frame or follow a track. Every clip on this page was made through the Pixel Dojo API.
Make your first P Video clip
Write a prompt above or start from any example on this page. One subscription covers every model on Pixel Dojo.
Start with P Video