13 answers · reviewed 18 November 2026
Pavo AI Video FAQ: 13 straight answers
No sign-up walls, no hidden pricing talk and no invented user quotes. These are the questions worth answering before you press Generate on a clip.
What is Pavo AI Video?
Pavo AI Video is a free AI video generator on pavoai.work. You start from one image (your own upload or one of our sample stills), describe the shot in a short prompt, choose a camera move and export a video clip. Rendering runs inside your browser, so there is no queue and no generation credit to spend.
Is Pavo AI Video free, and do I need an account?
Yes, it is free and there is no account. There is no registration, no trial, no credit system, no payment form and no watermark added to your export. Open the Create page, pick a still and render.
Are my images uploaded to a server?
No. Your image, prompt and finished clip stay on your device. The studio draws frames with the browser Canvas API and records them with MediaRecorder. Nothing is sent to pavoai.work or to any third-party AI service. The web server only sees a normal page request.
What file do I get, and how long is the clip?
You get a downloadable WebM video (VP9 or VP8, depending on your browser) at 30 fps in 16:9, 9:16 or 1:1. Choose a 3, 5 or 8 second duration. WebM plays in Chrome, Edge, Firefox and most modern editors; convert it to MP4 with any transcoder if your delivery pipeline requires it.
Is this the official Pavo product?
No. Pavo AI Video is an independent service operated on pavoai.work. It is not affiliated with, endorsed by or operated by pavoai.app, Agnes AI, CapCut, Dreamina or any third-party model provider. We do not provide account, billing or generation support for those products.
What are the 12 shot presets?
They are ready-made camera and timing setups for common shots: hard-light product reveals, wide exteriors, window-light portraits, macro splashes, rooftop aerials, studio paper motion, dialogue holds and more. Each preset lists a first-frame check, three action beats, identity anchors, a six-line prompt, two fixes and an editing handoff. Load one into the Create page and change one line at a time.
What is the six-line prompt format?
It is a short production brief made of SUBJECT, FRAME, ACTION, CAMERA, FIXED and END ON. Each line decides one thing, so when a clip fails you can change exactly one line and know what caused the fix. The free Prompt Builder writes it for you, and the Create page uses your prompt to suggest a camera move.
How does the prompt change the video?
The studio reads your prompt for motion words such as push, pull, pan, drift, hold and close-up, then suggests a matching camera path. You can accept the suggestion or override it. The six-line brief also becomes an optional caption overlay, so the exported clip carries its own shot description.
Will my character change face or grow extra fingers?
Not in Pavo AI Video. Because the renderer animates the pixels of the still you supplied, faces, hands and geometry stay exactly as they are — there is no generative redraw. Identity drift and extra fingers are failure modes of diffusion-based models; if you use one of those elsewhere, start from a still with a clear, front-facing subject and use our Render Fixes table to change one prompt line at a time.
Can I add captions or a logo to the exported clip?
You can toggle a caption bar that uses the first line of your prompt. We do not bake logos into the video: generated lettering is unreliable and a clean plate is much easier to title in your editor. The editing handoff on each preset says where to place type after export.
Which browsers support rendering?
Current Chrome, Edge and Firefox support Canvas captureStream and MediaRecorder, which the studio needs. Safari support for WebM recording varies by version; if recording is unavailable the studio tells you before starting instead of failing silently. Desktop browsers with hardware acceleration give the smoothest 30 fps capture.
Can I use the exported video commercially?
Yes, clips you render are yours. You are responsible for having the rights to the still you uploaded. The sample stills shipped with the studio are original artwork made for this project and may be used in your renders, including commercial work. See the Terms page for the details.
What are the limits of this renderer?
It produces motion from a single still — camera moves, subtle light sweeps and a held ending — not new objects, novel scenes or talking characters. If you need generative content synthesis you will need a diffusion-based model. We document the difference openly in How Rendering Works rather than pretending a local renderer is a large video model.
Where to go next
- 12 shot presets — start with a shot, not a blank box.
- Render fixes — six symptoms, six one-line edits.
- Five help guides — the reasoning behind the checks.
- About & independence — who maintains this and why.