> ## Documentation Index
> Fetch the complete documentation index at: https://docs.varg.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# OmniHuman

> ByteDance's full-body human animation from a single image and audio

## Overview

OmniHuman V1.5 (by ByteDance) generates full-body human animation videos from a single image and audio file. Unlike lipsync models that only animate the mouth, OmniHuman animates the entire body including gestures and posture.

| Model ID         | Input         | Resolution    | Credits | \~Cost |
| ---------------- | ------------- | ------------- | ------- | ------ |
| `omnihuman-v1.5` | Image + audio | 720p or 1080p | 100     | \$1.00 |

## Quick start

<CodeGroup>
  ```typescript SDK theme={null}
  import { createVarg } from "vargai/ai"

  const varg = createVarg({ apiKey: process.env.VARG_API_KEY! })

  const result = await varg.videoModel("omnihuman-v1.5").generate({
    imageUrl: "https://example.com/full-body-portrait.jpg",
    audioUrl: "https://example.com/speech.mp3",
  })

  console.log(result.video.url)
  ```

  ```bash cURL theme={null}
  curl -X POST https://api.varg.ai/v1/video \
    -H "Authorization: Bearer $VARG_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "omnihuman-v1.5",
      "prompt": "",
      "files": [
        { "url": "https://example.com/full-body-portrait.jpg" },
        { "url": "https://example.com/speech.mp3" }
      ]
    }'
  ```
</CodeGroup>

## Parameters

<ResponseField name="files" type="array" required>
  Two files: one image (full body preferred) and one audio.
</ResponseField>

<ResponseField name="provider_options" type="object">
  `resolution` — `"720p"` or `"1080p"` (default `"1080p"`). `turbo_mode` — boolean for faster generation at slightly lower quality.
</ResponseField>

## Pricing

| Model            | Credits | USD    |
| ---------------- | ------- | ------ |
| `omnihuman-v1.5` | 100     | \$1.00 |

## Tips

* **Full-body images work best.** OmniHuman animates hands, arms, and posture — use images showing the full or upper body.
* **Turbo mode** trades quality for speed. Good for previewing before final renders.
* **1080p resolution** is recommended for production content. 720p for faster iteration.
* **Quality is variable** — results depend heavily on the input image pose and composition.

## Related models

<CardGroup cols={2}>
  <Card title="VEED Fabric" icon="comment" href="/models/lipsync/veed-fabric">
    Faster and simpler. Face-only animation.
  </Card>

  <Card title="Sync V2 Pro" icon="lips" href="/models/lipsync/sync">
    Higher quality lip sync on existing video.
  </Card>
</CardGroup>