cinemasetfreeprunaai/p-video-2-pro-replace
Pruna's person-swap model. Put the person from a picture into a source video, keep the camera, motion, and soundtrack, from 1 to 120 seconds and 480p to 4K.
Capabilities
Cost
Community model (estimated from hardware time)
Input Parameters
| Name | Type | Description | Default | Constraints |
|---|---|---|---|---|
images* | array | Identity reference image(s) (1-2) to place into the video: one person, or two people, one picture each. | — | — |
video* | string(uri) | Source video (MP4, MOV or WebM, 1-120 s, up to 8K): motion, scene and audio source. Longer videos are cut to 120 s. | — | — |
enable_safety_filter | boolean | Check the inputs against the content policy. When off, only the always-on categories (sexual content involving minors, self-harm intent or instructions) are still blocked. | true | — |
ignore_audio | boolean | Ignore source audio during generation. If save_audio is true, the source audio is still saved with the final video. | false | — |
instruction_prompt | string | Further instruction on how to place the people from the reference images into the scene. | "" | — |
resolution | string | Output resolution (aspect ratio preserved); the output is 24 fps. 1080p, 2k and 4k generate at 768p and upscale it. | "768p" | 768p480p1080p2k4k |
save_audio | boolean | Save the video with audio. | true | — |
seed | integer | Random seed. Leave blank for random. | — | — |
imagesrequiredarrayIdentity reference image(s) (1-2) to place into the video: one person, or two people, one picture each.
videorequiredstringSource video (MP4, MOV or WebM, 1-120 s, up to 8K): motion, scene and audio source. Longer videos are cut to 120 s.
enable_safety_filterbooleanCheck the inputs against the content policy. When off, only the always-on categories (sexual content involving minors, self-harm intent or instructions) are still blocked.
trueignore_audiobooleanIgnore source audio during generation. If save_audio is true, the source audio is still saved with the final video.
falseinstruction_promptstringFurther instruction on how to place the people from the reference images into the scene.
""resolutionstringOutput resolution (aspect ratio preserved); the output is 24 fps. 1080p, 2k and 4k generate at 768p and upscale it.
"768p"save_audiobooleanSave the video with audio.
trueseedintegerRandom seed. Leave blank for random.
c0a4a1dafeabUpdated: 11/7/2026558 runs