p-video

P-Video generates videos using Pruna's optimized video models through the inference.sh CLI (the `belt` command), with Pruna's emphasis on speed without sacrificing quality. It exposes three models: P-Video (`pruna/p-video`) for text-to-video and image-to-video with audio support, WAN-T2V (`pruna/wan-t2v`) for text-to-video at 480p/720p, and WAN-I2V (`pruna/wan-i2v`) for animating still images at 480p/720p. The skill requires the belt CLI (installable via the belt CLI skill) and begins with `belt login`.

Videos are produced with `belt app run <app-id> --input '{...}'`. Text-to-video uses a `prompt` with optional `duration` and `resolution`; image-to-video adds an `image` URL plus a motion prompt. P-Video supports an `audio` input that syncs with the video, and a `draft` mode for faster, cheaper concept tests. WAN models take a `prompt`, `resolution`, and `duration`. Resolution and pricing vary by model: P-Video offers 720p and 1080p priced per second and varying by resolution and draft; WAN-T2V is $0.05 at 480p and $0.10 at 720p per video; WAN-I2V is $0.05 at 480p and $0.11 at 720p per video.

The skill documents examples for text-to-video, image-to-video, audio-driven generation, WAN text-to-video and image-to-video, 1080p high-quality output, and draft mode. Users can browse Pruna apps with `belt app list --namespace pruna` or `belt app store`. It links to documentation for running apps, streaming results for real-time progress, and a content-pipeline example for building media workflows, and references related skills for the full inference.sh platform, broader video generation, image-to-video, Pruna image generation, and text-to-speech for narration. The skill is scoped to the `Bash(belt *)` tool.