stable-diffusion-xl-lightning Beta

stable-diffusion-xl-lightning Beta

Text-to-Image • ByteDance • Hosted

@cf/bytedance/stable-diffusion-xl-lightning

SDXL-Lightning is a lightning-fast text-to-image generation model. It can generate high-quality 1024px images in a few steps.

Model Info
More information	link ↗
Beta	Yes
Unit Pricing	$0.00 per step

Usage

TypeScript

export interface Env {
  AI: Ai;
}

export default {
  async fetch(request, env): Promise<Response> {

    const inputs = {
      prompt: "cyberpunk cat",
    };

    const response = await env.AI.run(
      "@cf/bytedance/stable-diffusion-xl-lightning",
      inputs
    );

    return new Response(response, {
      headers: {
        "content-type": "image/jpg",
      },
    });
  },
} satisfies ExportedHandler<Env>;

Explain Code

curl

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run/@cf/bytedance/stable-diffusion-xl-lightning  \
  -X POST  \
  -H "Authorization: Bearer $CLOUDFLARE_API_TOKEN"  \
  -d '{ "prompt": "cyberpunk cat" }'

Parameters

Input

prompt

stringrequiredminLength: 1A text description of the image you want to generate

negative_prompt

stringText describing elements to avoid in the generated image

height

integerminimum: 256maximum: 2048The height of the generated image in pixels

width

integerminimum: 256maximum: 2048The width of the generated image in pixels

▶image[]

arrayFor use with img2img tasks. An array of integers that represent the image data constrained to 8-bit unsigned integer values

image_b64

stringFor use with img2img tasks. A base64-encoded string of the input image

▶mask[]

arrayAn array representing An array of integers that represent mask image data for inpainting constrained to 8-bit unsigned integer values

num_steps

integerdefault: 20maximum: 20The number of diffusion steps; higher values can improve quality but take longer

strength

numberdefault: 1A value between 0 and 1 indicating how strongly to apply the transformation during img2img tasks; lower values make the output closer to the input image

guidance

numberdefault: 7.5Controls how closely the generated image should adhere to the prompt; higher values make the image more aligned with the prompt

seed

integerRandom seed for reproducibility of the image generation

Output

The binding returns a ReadableStream with the output (check the model's output schema).

API Schemas

Input

{
  "type": "object",
  "properties": {
    "prompt": {
      "type": "string",
      "minLength": 1,
      "description": "A text description of the image you want to generate"
    },
    "negative_prompt": {
      "type": "string",
      "description": "Text describing elements to avoid in the generated image"
    },
    "height": {
      "type": "integer",
      "minimum": 256,
      "maximum": 2048,
      "description": "The height of the generated image in pixels"
    },
    "width": {
      "type": "integer",
      "minimum": 256,
      "maximum": 2048,
      "description": "The width of the generated image in pixels"
    },
    "image": {
      "type": "array",
      "description": "For use with img2img tasks. An array of integers that represent the image data constrained to 8-bit unsigned integer values",
      "items": {
        "type": "number",
        "description": "A value between 0 and 255"
      }
    },
    "image_b64": {
      "type": "string",
      "description": "For use with img2img tasks. A base64-encoded string of the input image"
    },
    "mask": {
      "type": "array",
      "description": "An array representing An array of integers that represent mask image data for inpainting constrained to 8-bit unsigned integer values",
      "items": {
        "type": "number",
        "description": "A value between 0 and 255"
      }
    },
    "num_steps": {
      "type": "integer",
      "default": 20,
      "maximum": 20,
      "description": "The number of diffusion steps; higher values can improve quality but take longer"
    },
    "strength": {
      "type": "number",
      "default": 1,
      "description": "A value between 0 and 1 indicating how strongly to apply the transformation during img2img tasks; lower values make the output closer to the input image"
    },
    "guidance": {
      "type": "number",
      "default": 7.5,
      "description": "Controls how closely the generated image should adhere to the prompt; higher values make the image more aligned with the prompt"
    },
    "seed": {
      "type": "integer",
      "description": "Random seed for reproducibility of the image generation"
    }
  },
  "required": [
    "prompt"
  ]
}

Explain Code

Output

{
  "type": "string",
  "contentType": "image/png",
  "format": "binary",
  "description": "The generated image in PNG format"
}