# Echomimic V3

EchoMimic V3 turns an image, audio, and text into a realistic talking avatar.

## API Information

- **Model Slug:** echomimic-v3
- **Branded URL:** https://www.eachlabs.ai/alibaba/echomimic/echomimic-v3
- **Provider:** Alibaba
- **Category:** Image to Video
- **Output Type:** video
- **Status:** active
- **Version:** 0.0.1
- **Base Cost:** Cost-per-second pricing: $0.20/sec of output video duration
- **Estimated Processing Time:** 280 seconds
- **Last Updated:** 2026-04-16
- **Interactive Demo:** https://www.eachlabs.ai/ai-models/echomimic-v3

## Pricing

- **Charge Type:** dynamic
- **Pricing Details:** Cost-per-second pricing: $0.20/sec of output video duration

### Pricing Rules

| Condition | Pricing |
| --- | --- |
| Rule 1 | Cost-per-second pricing: $0.20/sec of output video duration |

## Input Schema

| Parameter | Type | Required | Default | Constraints | Description |
|-----------|------|----------|---------|-------------|-------------|
| image_url | string | Yes | - | - | The URL of the image to use as a reference for the video generation. |
| audio_url | string | Yes | - | - | The URL of the audio to use as a reference for the video generation. |
| prompt | string | Yes | - | - | The prompt to use for the video generation. |
| negative_prompt | string | No | - | - | The negative prompt to use for the video generation. |
| num_frames_per_generation | integer | No | 121 | 49–161 | The number of frames to generate at once. |
| guidance_scale | number | No | 4.5 | 1–10 | The guidance scale to use for the video generation. |
| audio_guidance_scale | number | No | 2.5 | 0–10 | The audio guidance scale to use for the video generation. |
| seed | integer | No | - | - | The seed to use for the video generation. |

## 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": "echomimic-v3",
  "input": {
    "prompt": "A video of a person sitting calmly, speaking naturally with minimal movements. The background matches the reference image.",
    "audio_url": "https://storage.googleapis.com/magicpoint/inputs/echomimic-input-audio.mp3",
    "image_url": "https://storage.googleapis.com/magicpoint/inputs/echomimic-input-image.png"
  }
}'
```

## 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

**echomimic-v3 — Image-to-Video AI Model**

echomimic-v3, developed by Alibaba, is an image-to-video AI model designed to transform static images into dynamic talking avatars. By combining an image, audio input, and text guidance, echomimic-v3 generates realistic video output that synchronizes facial movements and expressions with provided audio. This approach solves a critical challenge in avatar creation: producing natural, lip-synced video content without requiring actors, studios, or extensive post-production work.

The model addresses the growing demand for personalized video content across marketing, education, and entertainment sectors. Rather than relying on pre-recorded footage or expensive video production, users can leverage echomimic-v3 to create talking avatars from a single image, making it accessible for creators and developers building AI-powered video generation applications.

## 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.