Lcm Lora Sdxl

🚀 潛在一致性模型 (LCM) LoRA：SDXL

潛在一致性模型 (LCM) LoRA 由 Simian Luo、Yiqin Tan、Suraj Patil、Daniel Gu 等人 在 LCM-LoRA: A universal Stable-Diffusion Acceleration Module 中提出。它是一個用於 stable-diffusion-xl-base-1.0 的蒸餾一致性適配器，可將推理步驟數減少至僅 2 - 8 步。

🚀 快速開始

LCM-LoRA 從 🤗 Hugging Face Diffusers 庫的 v0.23.0 版本開始支持。要運行該模型，首先需要安裝最新版本的 Diffusers 庫以及 peft、accelerate 和 transformers：

pip install --upgrade pip
pip install --upgrade diffusers transformers accelerate peft

⚠️ 重要提示

有關詳細的使用示例，建議查看官方 LCM-LoRA 文檔。

💻 使用示例

基礎用法

文本到圖像

適配器可以與其基礎模型 stabilityai/stable-diffusion-xl-base-1.0 一起加載。接下來，需要將調度器更改為 LCMScheduler，並且可以將推理步驟數減少到僅 2 到 8 步。請確保禁用 guidance_scale 或使用 1.0 到 2.0 之間的值。

import torch
from diffusers import LCMScheduler, AutoPipelineForText2Image

model_id = "stabilityai/stable-diffusion-xl-base-1.0"
adapter_id = "latent-consistency/lcm-lora-sdxl"

pipe = AutoPipelineForText2Image.from_pretrained(model_id, torch_dtype=torch.float16, variant="fp16")
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)
pipe.to("cuda")

# load and fuse lcm lora
pipe.load_lora_weights(adapter_id)
pipe.fuse_lora()

prompt = "Self-portrait oil painting, a beautiful cyborg with golden hair, 8k"

# disable guidance_scale by passing 0
image = pipe(prompt=prompt, num_inference_steps=4, guidance_scale=0).images[0]

圖像修復

LCM-LoRA 也可用於圖像修復。

import torch
from diffusers import AutoPipelineForInpainting, LCMScheduler
from diffusers.utils import load_image, make_image_grid

pipe = AutoPipelineForInpainting.from_pretrained(
    "diffusers/stable-diffusion-xl-1.0-inpainting-0.1",
    torch_dtype=torch.float16,
    variant="fp16",
).to("cuda")

# set scheduler
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)

# load LCM-LoRA
pipe.load_lora_weights("latent-consistency/lcm-lora-sdxl")
pipe.fuse_lora()

# load base and mask image
init_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/inpaint.png").resize((1024, 1024))
mask_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/inpaint_mask.png").resize((1024, 1024))

prompt = "a castle on top of a mountain, highly detailed, 8k"
generator = torch.manual_seed(42)
image = pipe(
    prompt=prompt,
    image=init_image,
    mask_image=mask_image,
    generator=generator,
    num_inference_steps=5,
    guidance_scale=4,
).images[0]
make_image_grid([init_image, mask_image, image], rows=1, cols=3)

高級用法

與風格化 LoRAs 結合使用

LCM-LoRA 可以與其他 LoRAs 結合使用，以在極少的步驟（4 - 8 步）內生成風格化圖像。在以下示例中，我們將 LCM-LoRA 與 papercut LoRA 結合使用。要了解更多關於如何結合 LoRAs 的信息，請參考 本指南。

import torch
from diffusers import DiffusionPipeline, LCMScheduler

pipe = DiffusionPipeline.from_pretrained(
    "stabilityai/stable-diffusion-xl-base-1.0",
    variant="fp16",
    torch_dtype=torch.float16
).to("cuda")

# set scheduler
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)

# load LoRAs
pipe.load_lora_weights("latent-consistency/lcm-lora-sdxl", adapter_name="lcm")
pipe.load_lora_weights("TheLastBen/Papercut_SDXL", weight_name="papercut.safetensors", adapter_name="papercut")

# Combine LoRAs
pipe.set_adapters(["lcm", "papercut"], adapter_weights=[1.0, 0.8])

prompt = "papercut, a cute fox"
generator = torch.manual_seed(0)
image = pipe(prompt, num_inference_steps=4, guidance_scale=1, generator=generator).images[0]
image

ControlNet

import torch
import cv2
import numpy as np
from PIL import Image

from diffusers import StableDiffusionXLControlNetPipeline, ControlNetModel, LCMScheduler
from diffusers.utils import load_image

image = load_image(
    "https://hf.co/datasets/huggingface/documentation-images/resolve/main/diffusers/input_image_vermeer.png"
).resize((1024, 1024))

image = np.array(image)

low_threshold = 100
high_threshold = 200

image = cv2.Canny(image, low_threshold, high_threshold)
image = image[:, :, None]
image = np.concatenate([image, image, image], axis=2)
canny_image = Image.fromarray(image)

controlnet = ControlNetModel.from_pretrained("diffusers/controlnet-canny-sdxl-1.0-small", torch_dtype=torch.float16, variant="fp16")
pipe = StableDiffusionXLControlNetPipeline.from_pretrained(
    "stabilityai/stable-diffusion-xl-base-1.0",
    controlnet=controlnet,
    torch_dtype=torch.float16,
    safety_checker=None,
    variant="fp16"
).to("cuda")

# set scheduler
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)

# load LCM-LoRA
pipe.load_lora_weights("latent-consistency/lcm-lora-sdxl")
pipe.fuse_lora()

generator = torch.manual_seed(0)
image = pipe(
    "picture of the mona lisa",
    image=canny_image,
    num_inference_steps=5,
    guidance_scale=1.5,
    controlnet_conditioning_scale=0.5,
    cross_attention_kwargs={"scale": 1},
    generator=generator,
).images[0]
make_image_grid([canny_image, image], rows=1, cols=2)

⚠️ 重要提示

此示例中的推理參數可能並非適用於所有情況，建議嘗試不同的 num_inference_steps、guidance_scale、controlnet_conditioning_scale 和 cross_attention_kwargs 參數值，並選擇最佳參數。

T2I Adapter

此示例展示瞭如何將 LCM-LoRA 與 Canny T2I-Adapter 和 SDXL 結合使用。

import torch
import cv2
import numpy as np
from PIL import Image

from diffusers import StableDiffusionXLAdapterPipeline, T2IAdapter, LCMScheduler
from diffusers.utils import load_image, make_image_grid

# Prepare image
# Detect the canny map in low resolution to avoid high-frequency details
image = load_image(
    "https://huggingface.co/Adapter/t2iadapter/resolve/main/figs_SDXLV1.0/org_canny.jpg"
).resize((384, 384))

image = np.array(image)

low_threshold = 100
high_threshold = 200

image = cv2.Canny(image, low_threshold, high_threshold)
image = image[:, :, None]
image = np.concatenate([image, image, image], axis=2)
canny_image = Image.fromarray(image).resize((1024, 1024))

# load adapter
adapter = T2IAdapter.from_pretrained("TencentARC/t2i-adapter-canny-sdxl-1.0", torch_dtype=torch.float16, varient="fp16").to("cuda")

pipe = StableDiffusionXLAdapterPipeline.from_pretrained(
    "stabilityai/stable-diffusion-xl-base-1.0", 
    adapter=adapter,
    torch_dtype=torch.float16,
    variant="fp16", 
).to("cuda")

# set scheduler
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)

# load LCM-LoRA
pipe.load_lora_weights("latent-consistency/lcm-lora-sdxl")

prompt = "Mystical fairy in real, magic, 4k picture, high quality"
negative_prompt = "extra digit, fewer digits, cropped, worst quality, low quality, glitch, deformed, mutated, ugly, disfigured"

generator = torch.manual_seed(0)
image = pipe(
    prompt=prompt,
    negative_prompt=negative_prompt,
    image=canny_image,
    num_inference_steps=4,
    guidance_scale=1.5, 
    adapter_conditioning_scale=0.8, 
    adapter_conditioning_factor=1,
    generator=generator,
).images[0]
make_image_grid([canny_image, image], rows=1, cols=2)

📄 許可證

本項目採用 openrail++ 許可證。