Pika's H3 Pro model behind one prompt bar: render from text, animate a single still, or direct the shot with up to nine images, three clips and three audio references — any length from 5 to 15 seconds, 480p or 768p, with native audio on every route.
Nothing attached renders from text alone. A single still becomes the first frame. Anything richer — more images, a clip, an audio track — runs as an ordered reference set. Same prompt, same settings, no mode switch to remember.
Every route renders dialogue, effects and ambience in sync with the motion, so a clip arrives ready to play rather than waiting on a score.
Up to nine images, three clips and three audio tracks in one run, each with a handle the prompt can point at. Mention something that is not attached and the run is held back until it matches.
Pick any whole second from 5 to 15 and any of eleven aspect ratios from 21:9 to 9:21 — or leave Auto and the first frame decides.
Any whole second from 5 to 15. Every route uses the same range, and length is priced per second of output.
What is attached decides. An empty attachment row renders from text alone; exactly one image becomes the first frame; anything more — extra images, a clip, an audio track — runs as an ordered reference set.
Up to nine images, three video clips and three audio tracks in one run, twelve in all. Clips run 2 to 15 seconds each and at most 15 seconds combined per kind; a set that runs over is flagged before you spend anything.
Yes. Pika H3 Pro renders its own soundtrack on the text, frame and reference routes alike, so there is no audio switch to set.
480p or 768p on every route. Aspect covers 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16 and 9:21, or Auto, which follows the first frame when there is one and 16:9 otherwise.
No. Reference audio rides alongside a reference image or clip — an audio-only set is held back before it is submitted.
