p-video-2-pro

Prerequisites

Install and load these skills before generating (skip if already in context via @pruna):

Or install the full suite once: npx skills add PrunaAI/pruna-skills@pruna -y

Follow each skill's Before generating / craft sections — do not restate guide content here.

Agent habit

Routing: `p-video-2-pro` is the standard cinematic generation clip (text / first frame / first+last frame) with generated audio. When imported image or video references should lock a subject or condition identity, style, motion, timing, or performance — or one uploaded audio track should be kept as the soundtrack (keep_audio) at 480p/768p — use `p-video-2-pro-reference`. Use `p-video-2` when the brief needs 1080p, an imported soundtrack, or draft previews. Use `p-video` for cheaper, faster drafts. Talking-head-only (no native scene audio) → `p-video-avatar`. Instruction edit of existing footage → `p-video-edit`.

In the first reply, name `p-video-2-pro` in backticks, confirm PRUNA_API_KEY (or stop with signup links from pruna-api), then ask for required inputs. Open intake → generation-diversity clarification intake before the first POST. When drafting motion prompts, follow Prompt craft (dynamic + faithful) and video-prompting (p-video-2-pro-prompting) — do not paste skill examples. Redirect when When NOT to use fits better.

Prompt craft (dynamic + faithful)

Every input.prompt must be fresh and specific, and must match the user's beat. Diversity never overrides the brief. Craft deltas vs p-video-2: video-prompting (p-video-2-pro-prompting). Write sound or dialogue into the prompt — this model has no audio upload.

Fidelity check (before pay): if you remove the user's named subject/action/setting from the prompt, the job is wrong — rewrite. Free axes only fill what the brief left open.

When showing a drafted prompt, still name `p-video-2-pro` (guides help craft; this tool owns the call).

Skill boundary

This skill = one p-video-2-pro prediction per invocation.

Out of scope (do not execute from this skill):

• Multi-scene assembly, concat, subagent orchestration, or parallel scene batches
• Imported image/video reference conditioning, with optional audio guidance or a kept soundtrack (keep_audio) → p-video-2-pro-reference
• Imported audio / mixed VO / 1080p / draft previews → p-video-2
• Motion transfer from a template video → p-video-animate
• Talking-head-only / lip-sync host without native scene audio → p-video-avatar
• Instruction edit of existing footage → p-video-edit

If the request exceeds one clip, stop and recommend: image-to-video (one beat), visual-transition-reel (multi-scene visual), or narrated-multi-scene (multi-scene + uploaded VO — that path uses p-video-2).

When NOT to use

Use a different skill instead:

HTTP (curl)

Upload for image-to-video / frame anchors

curl -X POST "https://api.pruna.ai/v1/files" \
  -H "apikey: ${PRUNA_API_KEY}" \
  -F "content=@/path/to/first-frame.png"

Pass urls.get as input.image (first frame) and/or input.last_frame_image (last frame). Do not upload audio for this model.

Create (async text-to-video — recommended)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-video-2-pro' \
  -d '{
    "input": {
      "prompt": "A sports car drifting through a neon-lit city at night, cinematic aerial shot",
      "duration": 5,
      "resolution": "768p",
      "aspect_ratio": "16:9"
    }
  }'

Poll and download: follow pruna-api.

Complete the random seed ritual from generation-diversity before writing prompts — do not pass the ritual string as API seed.

Billing is per returned second (not the requested duration). Rate limit: 250 requests per minute.

cost is the same output quality as speed, but cheaper and slower. prompt_upsampler does not change the price.

Image-to-video (first frame)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-video-2-pro' \
  -d '{
    "input": {
      "prompt": "The camera slowly pushes in, the person turns their head and smiles",
      "image": "https://api.pruna.ai/v1/files/FIRST_ID",
      "duration": 5,
      "resolution": "768p"
    }
  }'

First / last frame (visual transition)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-video-2-pro' \
  -d '{
    "input": {
      "prompt": "Start on the still product hero. The camera holds, then a slow push-in as light moves across the metal. End exactly on the last-frame packshot, label readable, no extra props.",
      "image": "https://api.pruna.ai/v1/files/START_ID",
      "last_frame_image": "https://api.pruna.ai/v1/files/END_ID",
      "duration": 8,
      "resolution": "768p",
      "mode": "speed",
      "prompt_upsampler": "turbo"
    }
  }'

There is no scene-anchor triple on this model (audio is not exposed). Uploaded VO / music → p-video-2.

Create (sync — quick test only)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H "apikey: ${PRUNA_API_KEY}" \
  -H 'Model: p-video-2-pro' \
  -H 'Try-Sync: true' \
  -d '{
    "input": {
      "prompt": "A sports car drifting through a neon-lit city at night, cinematic aerial shot",
      "duration": 5,
      "resolution": "480p",
      "mode": "cost"
    }
  }'

Before generating

1. Complete Prerequisites guide reading order (generation-diversity → video-prompting).
2. Ritual seed → draft a dynamic + faithful motion prompt (section above) → confirm mode (T2V / I2V / frame pair), duration (5–15s, default 5), resolution (480p / 768p), mode (cost / speed / quality), prompt_upsampler, and prompt with the user.
3. Pruna notes: when image or last_frame_image is set, aspect_ratio is ignored (canvas follows the stills). Output is 24 fps — do not send fps. Output includes generated audio; write sound or dialogue in the prompt. Do not send audio, draft, save_audio, or prompt_upsampling. Iterate in mode: cost when price is the priority (same quality as speed, but cheaper and slower) or mode: speed when wall-clock time matters (default); then final with mode: quality when fidelity matters. If the request is multi-scene — stop (see Skill boundary). Native 4K is not supported. 1080p / imported audio / draft → p-video-2.

Required input

• prompt (string)

Common optional fields

Do not send: audio, draft, fps, save_audio, prompt_upsampling.

Related

Related skills: