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

flux-depth-dev

flux-depth-dev

【效率与协作】Flux Depth Dev | Depth-Controlled Image Generation. Generate images guided by automatically extracted depth maps. Triggers: flux depth dev, depth control, depth map, image generation, controlnet depth

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: flux-depth-dev description: "Flux Depth Dev | Depth-Controlled Image Generation. Generate images guided by automatically extracted depth maps. Triggers: flux depth dev, depth control, depth map, image generation, controlnet depth" allowed-tools: Bash(curl *), WebFetch

Flux Depth Dev

Generate images using depth map control with the Flux Dev model. Automatically generates depth maps from control images to guide image generation with configurable output options.

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": "flux-depth-dev",
    "version": "0.0.1",
    "input": {
      "prompt": "a cozy cabin in a snowy mountain landscape, warm lights glowing from windows",
      "control_image": "https://example.com/landscape.jpg",
      "guidance": 10,
      "num_inference_steps": 28
    }
  }'

Parameters

ParameterTypeDefaultDescription
control_imagestring-Image used to control the generation. The depth map will be automatically generated
disable_safety_checkerbooleanfalseDisable safety checker for generated images
guidancenumber10Guidance for generated image
megapixelsstring1Resolution scale. enum: 1, 0.25
num_inference_stepsinteger28Number of denoising steps. Recommended range is 28-50
num_outputsinteger1Number of outputs to generate
output_formatstringwebpOutput format. enum: webp, jpg, png
output_qualityinteger80Quality when saving output images, from 0 to 100
promptstring-Prompt for generated image
seedinteger-Random seed. Set for reproducible generation

Examples

Interior design transformation:

curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "flux-depth-dev",
    "version": "0.0.1",
    "input": {
      "prompt": "a modern minimalist living room with Scandinavian furniture, natural light, white walls",
      "control_image": "https://example.com/room-photo.jpg",
      "guidance": 12,
      "num_inference_steps": 35,
      "output_format": "png",
      "output_quality": 95
    }
  }'

Multiple outputs with seed:

curl -X POST https://api.eachlabs.ai/v1/prediction \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "model": "flux-depth-dev",
    "version": "0.0.1",
    "input": {
      "prompt": "a medieval castle on a hill, dramatic storm clouds, dark fantasy style",
      "control_image": "https://example.com/hill-scene.jpg",
      "num_outputs": 2,
      "num_inference_steps": 40,
      "guidance": 8,
      "seed": 777
    }
  }'

Related Models

Documentation

我的收藏