prunaai/p-video-avatar 🔢🖼️✓❓🎥📝 → 🖼️

⭐ Official ▶️ 367.8K runs 📅 Mar 2026 ⚙️ Cog 0.18.0

audio-to-video lipsync text-to-video-with-audio

Performance

10.6sTypical run time

367.8KTotal runs

About

p-video-avatar is the fastest and cheapest avatar/lipsync video model on the market.

Example Output

Output

Performance Metrics

10.57s Prediction Time

10.58s Total Time

All Input Parameters

{
  "image": "https://replicate.delivery/pbxt/P0LFZw27zi6C5PnYjBEx4QS9h22dOIy3hwAUBAJ0n4YSUNil/replicate-prediction-qvbc943ybhrmt0cxtdtsd0gmsw.webp",
  "no_op": false,
  "voice": "Zephyr (Female)",
  "resolution": "720p",
  "video_prompt": "The person is talking.",
  "voice_prompt": "Say the following.",
  "voice_script": "p video avattar is now on Replicate!",
  "voice_language": "English (US)",
  "negative_prompt": "",
  "disable_safety_filter": true,
  "strength_negative_prompt": 0.5,
  "disable_prompt_upsampling": false
}

Input Parameters

seed Type: integer: Random seed. Set for reproducible generation.
audio Type: string: Optional uploaded audio to drive avatar speech. If provided, this is used instead of voice_script and voice settings.
image (required) Type: string: Input image (first frame). Supports jpg, jpeg, png, webp.
no_op Type: booleanDefault: false: Health check mode - returns status without inference.
voice Default: Zephyr (Female): Voice to use when generating speech from voice_script.
resolution Default: 720p: Resolution of the video.
video_prompt Type: stringDefault: The person is talking.: Optional visual prompt describing how the person should appear or behave while speaking.
voice_prompt Type: stringDefault: Say the following.: Optional style instructions for how to speak voice_script, such as tone, pacing, accent, or emotion. These instructions are not spoken.
voice_script Type: stringDefault:: Exact words the avatar should say. Required when no audio file is uploaded.
voice_language Default: English (US): Language/accent target for generated speech.
negative_prompt Type: stringDefault:: Disabled if empty.Mention what you do NOT want in the video, e.g. "subtitles, text, blurry, low quality, frames, watermark, titles, scene change". We recommend using multiple keywords at once.
disable_safety_filter Type: booleanDefault: true: Disable safety filter for prompts and input image. When disabled, prompts are not checked for unsafe content before generation.
strength_negative_prompt Type: numberDefault: 0.5Range: 0 - 4: Strength of the Negative Prompt. Optimal value can differ for different video lengths (Experimental Feature)
disable_prompt_upsampling Type: booleanDefault: false: When true, skip automatic enhancement of the visual video prompt and use video_prompt directly.

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

Video was trimmed to match audio length

Version Details

Version ID: ae9587aaba3cd72a7482c2a7230e028624d3bb098b0bff45df14a72ff3f1de3c
Version Created: October 11, 2026

Run on Replicate →