p-video-2-pro-reference

p-video-2-pro-reference generates a video from a prompt and at least one image or video reference. It is built on MiniMax H3. Audio references are optional. Resolutions are 480p and 768p. Duration is 5–15 seconds (default 5).

mode is speed (default) or cost. There is no quality mode. prompt_upsampler rewrites the prompt before generation: turbo (default, the fast upsampler), max, or off. It does not change the price.

keep_audio is off by default. Set it with exactly one audio file to keep that file as the video's audio instead of only using it as a reference. The video is generated to match it. Shorter audio plays first and the model generates the rest; longer audio is cut to duration. An audio file within 1 second of duration sets the video length. It does not change the price.

first_frame_image and last_frame_image are optional soft keyframes. They count as reference images. A start or end frame needs prompt upsampling, so prompt_upsampler=off still runs turbo when one of those frames is set.

Reference limits, enforced by the model: up to 9 images including start and end frames (20 MiB, 4096 px), 3 videos (100 MiB, 1080p, 60 fps, 2–15 seconds each, 15 seconds combined), and 3 audio files (50 MiB, 15 seconds combined).

Rate Limit: 250 requests per minute

Category: Video Generation

Pricing: billed from the finished run. Output seconds use a rate for resolution and mode. Reference video seconds use a rate for the same pair. Each reference image has a per-image rate, and reference audio has a per-second rate. Start and end frames count as images. Image and audio rates do not depend on resolution.

Output	Price
480p, mode=cost	$0.015 per second
768p, mode=cost	$0.025 per second
480p, mode=speed	$0.02 per second
768p, mode=speed	$0.035 per second

Reference video	Price
480p, mode=cost	$0.03 per second
768p, mode=cost	$0.06 per second
480p, mode=speed	$0.04 per second
768p, mode=speed	$0.06 per second

Reference images	Price
mode=cost	$0.005 each
mode=speed	$0.01 each

Reference audio	Price
mode=cost	$0.0015 per second
mode=speed	$0.0025 per second

Output length and reference usage are measured from the finished run, so the final price is known once generation completes. Omitted mode is billed as speed. prompt_upsampler and keep_audio do not change the price.

Output seconds are the wrapper's whole-second count, int(num_frames / 24). A 5 second request is 124 frames and bills as 5 seconds. A 15 second request is 345 frames and bills as 14 seconds. Each reference clip counts only up to that output length plus 0.5 seconds, and reference video is capped at 15 seconds combined. A 15 second reference clip on a 5 second output bills about 5.5 reference-video seconds, not the full upload. The per-second rates above are with no references. The shortest run is 5 seconds, so the cheapest request is $0.075 (cost, 480p, no references).

Quickstart

Image reference (asynchronous)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H 'apikey: YOUR_API_KEY' \
  -H 'Model: p-video-2-pro-reference' \
  -d '{
    "input": {
      "prompt": "The person in the reference walks through a sunlit market",
      "image": ["https://api.pruna.ai/v1/files/fqadqq42xq"],
      "duration": 5,
      "resolution": "768p",
      "aspect_ratio": "adaptive"
    }
  }'

Cost mode (synchronous)

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H 'apikey: YOUR_API_KEY' \
  -H 'Model: p-video-2-pro-reference' \
  -H 'Try-Sync: true' \
  -d '{
    "input": {
      "prompt": "The person in the reference walks through a sunlit market",
      "image": ["https://api.pruna.ai/v1/files/fqadqq42xq"],
      "duration": 5,
      "resolution": "480p",
      "mode": "cost"
    }
  }'

Video, audio, and soft keyframes

Upload each file first, then pass the returned URLs:

curl -X POST "https://api.pruna.ai/v1/files" \
  -H "apikey: YOUR_API_KEY" \
  -F "content=@/path/to/your/file.jpg"

curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H 'apikey: YOUR_API_KEY' \
  -H 'Model: p-video-2-pro-reference' \
  -d '{
    "input": {
      "prompt": "Continue the motion and match the reference audio",
      "video": ["https://api.pruna.ai/v1/files/video123"],
      "audio": ["https://api.pruna.ai/v1/files/audio123"],
      "first_frame_image": "https://api.pruna.ai/v1/files/start123",
      "last_frame_image": "https://api.pruna.ai/v1/files/end123",
      "duration": 5,
      "resolution": "768p",
      "mode": "speed"
    }
  }'

Parameters

Required Parameters

Parameter	Type	Description
prompt	string	Text prompt for video generation

Send at least one image or video reference with the prompt. Audio is optional.

Optional Parameters

Parameter	Type	Default	Description
prompt_upsampler	string	"turbo"	Rewrite the prompt before generation: "turbo" (fast default), "max", or "off". Independent of mode
image	array of URIs	—	Reference images
video	array of URIs	—	Reference videos
audio	array of URIs	—	Reference audio. Set keep_audio to use one file as the video's audio
keep_audio	boolean	false	Keep the uploaded audio as the video's audio. Needs exactly one audio file
first_frame_image	string (URI)	—	Optional opening image. Soft keyframe; counts as a reference image
last_frame_image	string (URI)	—	Optional closing image. Soft keyframe; counts as a reference image
duration	integer	5	Video duration in seconds (5-15). With keep_audio, audio within 1 s of this sets the length
resolution	string	"768p"	"480p" or "768p"
mode	string	"speed"	"speed" or "cost"
aspect_ratio	string	"16:9"	"adaptive", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3", "1:1". adaptive follows the references
seed	integer	random	Random seed for reproducible generation

Use is subject to the MiniMax H3 Community License, including its Acceptable Use Policy. See Terms, section 7.2.

License: https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/LICENSE