Pruna AI P Video 2 Pro Image To Video API Documentation

Pruna AI P Video 2 Pro Image To Video API Documentation

Playground

Pruna P-Video-2-Pro Image-to-Video generates videos from input images with optional last-frame guidance, 480P / 768P output, generated audio, and explicit controls for duration, resolution, and draft mode. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Pruna P-Video-2 Pro Image-to-Video generates videos from a first-frame reference image and a text prompt. The input image determines the output canvas, while the prompt controls motion, scene direction, camera movement, and visual style. You can also provide an optional last-frame image for stronger start-to-end-frame control.

Why Choose This?

• Image-to-video generation
  Animate a first-frame image into a generated video.

• First-frame canvas control
  The input image determines the output canvas and visual starting point.

• Optional last-frame guidance
  Use last_image to guide the ending frame of the generated video.

• 480p and 768p output
  Use 480p for lower-cost generation or 768p for higher-resolution output.

• Speed and quality modes
  Choose mode=speed for faster generation or mode=quality for higher-quality generation. Default: speed.

Parameters

How to Use

1. Write a prompt — Describe the motion, action, camera movement, lighting, mood, and scene direction.
2. Upload a first-frame image — Provide the image that should define the starting frame and output canvas.
3. Add last-frame guidance optional — Use last_image when you want to guide the final frame.
4. Set duration — Choose a video length from 5 to 15 seconds.
5. Choose resolution — Use 480p for lower-cost generation or 768p for higher-resolution output.
6. Choose generation mode — Select speed for faster generation or quality for higher-quality generation. Pricing depends on the selected mode and resolution.
7. Set prompt upsampling — Use off, turbo, or max depending on how much prompt expansion you want.
8. Set seed optional — Use a fixed seed for reproducible results, or omit it for random generation.
9. Submit — Generate the final image-to-video output.

Pricing

Pricing is based on the requested duration, selected resolution, and generation mode.

Example Costs

The default settings are 768p, speed, and 5 seconds, costing $0.175 per request.

Billing uses the requested duration, not the measured duration of the generated output. Generated audio is included. Prompt upsampling, seed selection, and supported image-conditioning inputs do not add separate charges.

Best Use Cases

• Image animation — Turn a still image into a generated video.
• Start-and-end-frame generation — Use image and last_image to guide both ends of the video.
• Product and creative videos — Animate product images, campaign visuals, posters, or concept frames.
• Character motion — Add motion, expression, or camera movement to a character or portrait image.
• Creative prototyping — Test motion direction, pacing, and visual style from a fixed first frame.
• Prompt iteration — Compare speed and quality generation modes and independently adjust prompt upsampling.

Pro Tips

• Use a clear, high-quality first-frame image for stronger subject and composition control.
• Add last_image when the final pose, framing, or scene state matters.
• Write prompts that describe visible motion, not just visual style.
• Include subject action, camera movement, lighting, mood, and scene progression.
• Use mode=speed for faster, lower-cost iteration, or mode=quality when prioritizing output quality.
• Use prompt_upsampler=off when you want the model to follow your original prompt more directly.
• Use prompt_upsampler=max when the prompt is short and needs stronger expansion.
• Use 480p for lower-cost tests and 768p for higher-resolution output.
• Set a fixed seed when comparing prompt or parameter changes.

Notes

• prompt and image are required.
• duration supports values from 5 to 15 seconds.
• The first-frame image determines the output canvas.
• last_image is optional and is used for ending-frame guidance.
• mode and prompt_upsampler are independent settings. mode controls the generation recipe and price; prompt_upsampler controls prompt expansion without changing the price.

Related Models

• Pruna P-Video-2 Pro Text-to-Video — Generate Pro videos directly from text prompts.
• Pruna P-Video-2 Pro Image-to-Video — Generate Pro videos from a first-frame image and motion prompt.
• Pruna P-Video-2 Text-to-Video — Generate videos directly from text prompts.
• Pruna P-Video-2 Image-to-Video — Generate videos from an input image and motion prompt.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "duration": 5,
  "resolution": "768p",
  "mode": "speed",
  "prompt_upsampler": "turbo"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/pruna-ai/p-video-2-pro/image-to-video" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters