[Integrations](https://spiral.ai/integrations) › [Amp](https://spiral.ai/integrations/amp) › [OmniHuman](https://spiral.ai/integrations/amp/omnihuman)

# How to Use OmniHuman in Amp

OmniHuman turns a character image and recorded audio into a speaking or singing video with synchronized expressions and gestures. This original OmniHuman endpoint suits short presenter clips and character performances, with audio strictly under 30 seconds.

|  |  |
| --- | --- |
| Model | OmniHuman |
| Available through | fal |
| Category | Video |
| Price | Up to $10.92 per run at the default inputs for 60 seconds of input. |
| Returns | video |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with OmniHuman in Amp

- Use OmniHuman to animate assets/presenter.png with assets/welcome.wav for our onboarding screen. Check that the recording is under 30 seconds, then save the generated clip in the project's media folder.
- Use OmniHuman to preview the three recorded NPC lines in assets/dialogue with assets/characters/guide.png. Check that each recording is under 30 seconds and label each output with its dialogue filename.
- Use OmniHuman to turn assets/mascot.png and assets/release-announcement.wav into a release announcement clip. Check the audio length first and keep the result with the release media assets.
- Use OmniHuman with assets/singer.png and the first 20 seconds of assets/vocal-demo.wav. Save the singing performance as a preview for the music scene.
- Use OmniHuman to compare assets/voice/calm.wav and assets/voice/energetic.wav against the same presenter portrait. Confirm both recordings are under 30 seconds and create a local review page showing the two takes.

## How to set up OmniHuman in Amp

Pick the way that suits you. Each takes about two minutes.

### Use Spiral's MCP server

1. **Add the server**: No header and no client ID or secret: Amp registers itself with Spiral.

```bash
amp mcp add spiral https://api.spiral.ai/mcp
```

2. **Sign in**: Start `amp`. Amp opens Spiral's sign-in page in your browser: sign in, pick the organization and click Authorize. No API key.

3. **Check it**

```bash
amp mcp doctor
```

### Use the Spiral skill

1. **Send Amp this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run OmniHuman**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/bytedance/omnihuman",
    "input": {
      "image_url": "https://storage.googleapis.com/falserverless/example_inputs/omnihuman.png",
      "audio_url": "https://storage.googleapis.com/falserverless/example_inputs/omnihuman_audio.mp3"
    }
  }'
```

Then paste one of the prompts above: Amp finds OmniHuman on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/amp

## OmniHuman tools and prices

- [OmniHuman](https://spiral.ai/tools/fal/fal-ai/bytedance/omnihuman) (fal): $0.182 per second.

## What can you do with OmniHuman in Amp?

Create presenter clips and character dialogue previews from assets already in your project. Have the coding agent check recording lengths and organize generated takes for review.

### Product introduction

Animate a presenter image with a recorded product introduction under 30 seconds. Use the clip as the spoken opening before separately edited product footage.

### One-concept lesson

Pair an instructor portrait with a short explanation of one concept. Build a longer lesson from separately reviewed clips.

### Game dialogue preview

Combine character artwork with a recorded line to preview facial performance and gestures before committing to a scene.

### Singing character excerpt

Animate a character image with a vocal excerpt under 30 seconds for a song teaser or performance study.

### Mascot announcement

Give an illustrated mascot a short recorded announcement for an event, release, or seasonal update.

### Performance comparison

Use the same portrait with two recordings of a line to compare how different deliveries affect the generated expressions and movements.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All Amp guides](https://spiral.ai/integrations/amp)

## OmniHuman in other agents

- [ChatGPT](https://spiral.ai/integrations/chatgpt/omnihuman)
- [Claude](https://spiral.ai/integrations/claude-desktop/omnihuman)
- [Claude Code](https://spiral.ai/integrations/claude-code/omnihuman)
- [Cline](https://spiral.ai/integrations/cline/omnihuman)
- [Codex](https://spiral.ai/integrations/codex/omnihuman)
- [Cursor](https://spiral.ai/integrations/cursor/omnihuman)
- [Devin](https://spiral.ai/integrations/devin/omnihuman)
- [dots](https://spiral.ai/integrations/dots/omnihuman)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/omnihuman)
- [Gemini](https://spiral.ai/integrations/gemini/omnihuman)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/omnihuman)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/omnihuman)
- [goose](https://spiral.ai/integrations/goose/omnihuman)
- [Grok](https://spiral.ai/integrations/grok/omnihuman)
- [Grok Build](https://spiral.ai/integrations/grok-build/omnihuman)
- [Hermes](https://spiral.ai/integrations/hermes/omnihuman)
- [Instinct](https://spiral.ai/integrations/instinct/omnihuman)
- [Manus](https://spiral.ai/integrations/manus/omnihuman)
- [Muse](https://spiral.ai/integrations/muse/omnihuman)
- [Muse Code](https://spiral.ai/integrations/muse-code/omnihuman)
- [n8n](https://spiral.ai/integrations/n8n/omnihuman)
- [Notion](https://spiral.ai/integrations/notion/omnihuman)
- [OpenClaw](https://spiral.ai/integrations/openclaw/omnihuman)
- [opencode](https://spiral.ai/integrations/opencode/omnihuman)
- [Perplexity](https://spiral.ai/integrations/perplexity/omnihuman)
- [Raycast](https://spiral.ai/integrations/raycast/omnihuman)
- [VS Code](https://spiral.ai/integrations/vscode/omnihuman)
- [Windsurf](https://spiral.ai/integrations/windsurf/omnihuman)
- [Zapier](https://spiral.ai/integrations/zapier/omnihuman)

## More Amp guides

- [Mureka 9.5 Lyrics Video](https://spiral.ai/integrations/amp/mureka-9-5-lyrics-video)
- [OmniHuman 1.5](https://spiral.ai/integrations/amp/omnihuman-1-5)
- [Mirelo SFX 2.0 Video to Video](https://spiral.ai/integrations/amp/mirelo-sfx-2-0-video-to-video)
- [OmniHuman 2.0](https://spiral.ai/integrations/amp/omnihuman-2-0)
- [MiniMax Video 01 Text to Video](https://spiral.ai/integrations/amp/minimax-video-01-text-to-video)
- [Ovi Image to Video](https://spiral.ai/integrations/amp/ovi-image-to-video)

## Questions about OmniHuman in Amp

**Can Amp use OmniHuman?**

Yes. Add Spiral to Amp as an MCP server, and Amp can find and run OmniHuman like any built-in tool.

**How much does OmniHuman cost in Amp?**

Up to $10.92 per run at the default inputs for 60 seconds of input. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Amp plan is separate.

**Do I need a fal account to use OmniHuman in Amp?**

No. Spiral runs OmniHuman on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up OmniHuman in Amp?**

About two minutes: add the server, then sign in, then check it.

**Does Amp wait while OmniHuman runs?**

OmniHuman runs in the background. Amp starts the run, checks on it and gives you the video when it's ready.

**What is OmniHuman best used for?**

OmniHuman suits short presenter clips and recorded character performances where speech, expressions, and gestures need to work together. Start with a character image and the finished recording.

**How long can an OmniHuman clip be?**

Keep the audio strictly under 30 seconds. The generated video follows the recording's duration, so divide longer material into separate takes.

**Does OmniHuman animate more than a face?**

Yes. OmniHuman supports close-up portraits, half-body images, and full-body images, allowing gestures to appear alongside facial animation.

**Can OmniHuman make a character sing?**

Yes. Singing is supported alongside speech; use a recorded vocal excerpt under 30 seconds for this endpoint.

**Can OmniHuman animate illustrated characters?**

Yes. The research demonstrations include cartoons, animals, and artificial objects. Try a short sample with the intended artwork before producing a series.

