# Infinitalk | Video to Video

InfiniTalk Video-to-Video enables advanced video-to-video transformation by synchronizing visual content with spoken audio. It transfers speech-driven expressions, lip movements, and facial dynamics from a source video to a target video, delivering natural, high-fidelity results with smooth motion and realistic playback. Ideal for dubbing, avatar animation, and multilingual video generation workflows.

## API Information

- **Model Slug:** infinitalk-video-to-video
- **Branded URL:** https://www.eachlabs.ai/infinitetalk/infinitetalk/infinitalk-video-to-video
- **Provider:** infinitetalk
- **Category:** Video to Video
- **Output Type:** video
- **Status:** active
- **Base Cost:** $0.03–$0.06 per second, depending on resolution
- **Estimated Processing Time:** 270 seconds
- **Interactive Demo:** https://www.eachlabs.ai/ai-models/infinitalk-video-to-video

## Pricing

- **Charge Type:** dynamic
- **Estimate:** $0.03–$0.06 per second, depending on resolution
- **Pricing Details:** 720p: $0.06
- **Pricing Details:** default: $0.03

## Input Schema

| Parameter | Type | Required | Default | Constraints | Description |
|-----------|------|----------|---------|-------------|-------------|
| audio | string | Yes | - | - | Audio URL to drive the avatar animation. Supports speech or singing audio to animate the person in the video. |
| video | string | Yes | - | - | Video URL of the person to animate. The person's face and body movements will be driven by the provided audio. |
| mask_image | string | No | - | - | Optional mask image to specify which regions can move. Do not upload the full image as mask — only cover the regions you want to animate. |
| prompt | string | No | - | - | Optional text prompt to guide expression, style, or pose. Keep prompts concise — too many instructions can produce noisy videos. |
| resolution | string | No | 480p | 480p, 720p | Output video resolution. 480p: faster generation, lower cost. 720p: higher quality, higher cost. |
| seed | integer | No | 1 | - | Random seed for reproducibility. -1 for random generation, or set a fixed number to reproduce the same result. |

## Example Request

```bash
curl -X POST https://api.eachlabs.ai/v1/prediction/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "infinitalk-video-to-video",
  "input": {
    "audio": "https://cdn-us.eachlabs.ai/defaults/93803cb34e1cf50e438012c2449c93d67e5fac3f4a1f9d28bfef22996c3c71c3.mp3",
    "video": "https://cdn-us.eachlabs.ai/defaults/fdc9b332565f0b2927c7a7b9a0622858f53d9a498c281ff0bbee0eea60f84e65.mp4"
  }
}'
```

## Output Schema

Response returned by `GET /v1/prediction/{id}` when the job completes:

```json
{
  "status": "success",
  "predictionID": "string",
  "output": "string (URL of generated video)",
  "metrics": {
    "predict_time": "number (seconds)"
  }
}
```

## Polling

```bash
curl https://api.eachlabs.ai/v1/prediction/{PREDICTION_ID} \
  -H "Authorization: Bearer YOUR_API_KEY"
```

| Status | Meaning |
|--------|---------|
| `processing` | Still running — poll again |
| `success` | Done — read `output` |
| `error` | Failed — read `message` / `details` |

## Webhook (alternative to polling)

Pass `"webhook_url": "https://your.host/path"` in the create request. Eachlabs POSTs this payload when the job ends:

```json
{
  "exec_id": "prediction-uuid",
  "status": "succeeded",
  "output": "https://...",
  "error": ""
}
```

`status` is `"succeeded"` or `"failed"`. `exec_id` equals the `predictionID` from create. Return 2xx within 30 seconds.

## Errors

Error body: `{ "status": "error", "message": "...", "details": "..." }`

| Code | Meaning |
|------|---------|
| `400` | Invalid input |
| `401` | Missing / invalid `Authorization` bearer token |
| `404` | Unknown model or prediction id |
| `429` | Rate limit — 100 creates / min, 10 concurrent per key |
| `5xx` | Retry with backoff |

## Overview

**infinitalk-video-to-video — Video-to-Video AI Model**

infinitalk-video-to-video, developed by infinitetalk as part of the InfiniteTalk family, revolutionizes **video-to-video AI model** workflows by transferring precise lip-sync, facial expressions, head motion, and full-body gestures from a source video to a target video driven by new audio. This enables seamless dubbing and avatar animation for infinite-length talking videos without breaking character consistency, solving the challenge of creating natural, multilingual content from existing footage. Ideal for creators seeking **infinitetalk video-to-video** capabilities, it supports long-form outputs up to minutes or more, maintaining stable identity and realistic dynamics throughout.

## Usage Notes

- API Base URL: `https://api.eachlabs.ai/v1`
- Authentication: send `Authorization: Bearer YOUR_API_KEY`. Generate a key from the Eachlabs dashboard at https://www.eachlabs.ai/dashboard/api-keys.
- File-typed parameters (`*_url`, `image_url`, `video_url`, `audio_url`, etc.) accept publicly-reachable HTTPS URLs only. Upload your asset first (GCS / S3 / your CDN) and pass the resulting URL. Data-URIs and localhost URLs are rejected.
- For structured parameters (arrays / objects) send real JSON values, not stringified payloads.
- Monetary values are reported in USD; per-token / per-megapixel rates may be billed in micro-cents internally.
- Prefer `webhook_url` over polling for long-running predictions — see the Webhook Callback section.