界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
效率与协作 · SKILL.md

vidu-q2-reference-to-image

vidu-q2-reference-to-image

【效率与协作】Vidu Q2 | Reference to Image. Generate images using reference images for consistent subject appearance. Triggers: reference to image, vidu, consistent character, image generation

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: vidu-q2-reference-to-image description: "Vidu Q2 | Reference to Image. Generate images using reference images for consistent subject appearance. Triggers: reference to image, vidu, consistent character, image generation" allowed-tools: Bash(curl *), WebFetch

Vidu Q2 | Reference to Image

Generate images using reference images to maintain consistent subject appearance across generations. Combine text prompts with visual references for precise control.

Quick Start

Requires an each::labs API key. Get one at eachlabs.ai.

curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "vidu-q2-reference-to-image",
    "version": "0.0.1",
    "input": {
      "prompt": "the character standing in front of a castle, heroic pose",
      "reference_image_urls": ["https://example.com/character-ref.jpg"],
      "aspect_ratio": "16:9"
    }
  }'

Parameters

ParameterTypeDefaultDescription
aspect_ratiostring16:9The aspect ratio of the output image. enum: 16:9, 9:16, 1:1
promptstringText prompt for image generation, max 1500 characters.
reference_image_urlsarrayURLs of the reference images to use for consistent subject appearance.
seedintegerRandom seed for generation.

Examples

Character in new scene:

curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "vidu-q2-reference-to-image",
    "version": "0.0.1",
    "input": {
      "prompt": "the person sitting in a modern coffee shop, reading a book",
      "reference_image_urls": ["https://example.com/person-ref.jpg"],
      "aspect_ratio": "1:1",
      "seed": 42
    }
  }'

Multiple references for consistency:

curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "vidu-q2-reference-to-image",
    "version": "0.0.1",
    "input": {
      "prompt": "the character wearing a spacesuit, walking on Mars surface",
      "reference_image_urls": ["https://example.com/char-front.jpg", "https://example.com/char-side.jpg"],
      "aspect_ratio": "16:9"
    }
  }'

Related Models

Documentation

我的收藏