TCD SDXL LoRA

🚀 Trajectory Consistency Distillation

Official Model Repo of the paper Trajectory Consistency Distillation, offering a novel distillation technology to transfer knowledge from pre - trained diffusion models to a few - step sampler.

Project Information

🚀 Quick Start

For more information, please check the GitHub Repo and Project Page. Also welcome to try the demo hosted on 🤗 Space.

✨ Features

A Solemn Statement Regarding the Plagiarism Allegations

We regret to hear about the serious accusations from the CTM team.

We sadly found out our CTM paper (ICLR24) was plagiarized by TCD! It's unbelievable😢—they not only stole our idea of trajectory consistency but also comitted "verbatim plagiarism," literally copying our proofs word for word! Please help me spread this. pic.twitter.com/aR6pRjhj5X

— Dongjun Kim (@gimdong58085414) March 25, 2024

Before this post, we already had several rounds of communication with CTM's authors. We shall proceed to elucidate the situation here.

We regret to hear about the serious accusations from the CTM team @gimdong58085414. I shall proceed to elucidate the situation and make an archive here. We already have several rounds of communication with CTM's authors. https://t.co/BKn3w1jXuh

— Michael (@Merci0318) March 26, 2024

1. In the first arXiv version, we provided citations and discussion in A. Related Works:

   Kim et al. (2023) proposes a universal framework for CMs and DMs. The core design is similar to ours, with the main differences being that we focus on reducing error in CMs, subtly leverage the semi - linear structure of the PF ODE for parameterization, and avoid the need for adversarial training.

2. In the first arXiv version, we indicated in D.3 Proof of Theorem 4.2

   In this section, our derivation mainly borrows the proof from (Kim et al., 2023; Chen et al., 2022).

   And we have never intended to claim credits. As we mentioned in our email, we would like to extend a formal apology to the CTM authors for the clearly inadequate level of referencing in our paper. We will provide more credits in the revised manuscript.

3. In the updated second arXiv version, we expanded our discussion to elucidate the relationship with the CTM framework. Additionally, we removed some proofs that were previously included for completeness.

4. CTM and TCD are different from motivation, method to experiments. TCD is founded on the principles of the Latent Consistency Model (LCM), aimed to design an effective consistency function by utilizing the exponential integrators.

5. The experimental results also cannot be obtained from any type of CTM algorithm.

   5.1 Here we provide a simple method to check: use our sampler here to sample the checkpoint CTM released, or vice versa.

   5.2 CTM also provided a training script. We welcome anyone to reproduce the experiments on SDXL or LDM based on the CTM algorithm.

We believe the assertion of plagiarism is not only severe but also detrimental to the academic integrity of the involved parties. We earnestly hope that everyone involved gains a more comprehensive understanding of this matter.

Introduction

TCD, inspired by Consistency Models, is a novel distillation technology that enables the distillation of knowledge from pre - trained diffusion models into a few - step sampler. In this repository, we release the inference code and our model named TCD - SDXL, which is distilled from SDXL Base 1.0. We provide the LoRA checkpoint in this repository.

✨TCD has the following advantages:

• Flexible NFEs: For TCD, the NFEs can be varied at will (compared with Turbo), without adversely affecting the quality of the results (compared with LCMs), where LCM experiences a notable decline in quality at high NFEs.
• Better than Teacher: TCD maintains superior generative quality at high NFEs, even exceeding the performance of DPM - Solver++(2S) with origin SDXL. It is worth noting that there is no additional discriminator or LPIPS supervision included during training.
• Freely Change the Detailing: During inference, the level of detail in the image can be simply modified by adjusting one hyper - parameter gamma. This option does not require the introduction of any additional parameters.
• Versatility: Integrated with LoRA technology, TCD can be directly applied to various models (including the custom Community Models, styled LoRA, ControlNet, IP - Adapter) that share the same backbone, as demonstrated in the Usage.
• Avoiding Mode Collapse: TCD achieves few - step generation without the need for adversarial training, thus circumventing mode collapse caused by the GAN objective. In contrast to the concurrent work [SDXL - Lightning](https://huggingface.co/ByteDance/SDXL - Lightning), which relies on Adversarial Diffusion Distillation, TCD can synthesize results that are more realistic and slightly more diverse, without the presence of "Janus" artifacts.

For more information, please refer to our paper Trajectory Consistency Distillation.

📦 Installation

To run the model yourself, you can leverage the 🧨 Diffusers library.

pip install diffusers transformers accelerate peft

And then clone the repo.

git clone https://github.com/jabir-zheng/TCD.git
cd TCD

💻 Usage Examples

Here, we demonstrate the applicability of our TCD LoRA to various models, including SDXL, SDXL Inpainting, a community model named Animagine XL, a styled LoRA Papercut, pretrained Depth Controlnet, Canny Controlnet and [IP - Adapter](https://github.com/tencent-ailab/IP - Adapter) to accelerate image generation with high quality in few steps.

Basic Usage

Text - to - Image generation

import torch
from diffusers import StableDiffusionXLPipeline
from scheduling_tcd import TCDScheduler 

device = "cuda"
base_model_id = "stabilityai/stable-diffusion-xl-base-1.0"
tcd_lora_id = "h1t/TCD-SDXL-LoRA"

pipe = StableDiffusionXLPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16, variant="fp16").to(device)
pipe.scheduler = TCDScheduler.from_config(pipe.scheduler.config)

pipe.load_lora_weights(tcd_lora_id)
pipe.fuse_lora()

prompt = "Beautiful woman, bubblegum pink, lemon yellow, minty blue, futuristic, high-detail, epic composition, watercolor."

image = pipe(
    prompt=prompt,
    num_inference_steps=4,
    guidance_scale=0,
    # Eta (referred to as `gamma` in the paper) is used to control the stochasticity in every step.
    # A value of 0.3 often yields good results.
    # We recommend using a higher eta when increasing the number of inference steps.
    eta=0.3, 
    generator=torch.Generator(device=device).manual_seed(0),
).images[0]

Advanced Usage

Inpainting

import torch
from diffusers import AutoPipelineForInpainting
from diffusers.utils import load_image, make_image_grid
from scheduling_tcd import TCDScheduler 

device = "cuda"
base_model_id = "diffusers/stable-diffusion-xl-1.0-inpainting-0.1"
tcd_lora_id = "h1t/TCD-SDXL-LoRA"

pipe = AutoPipelineForInpainting.from_pretrained(base_model_id, torch_dtype=torch.float16, variant="fp16").to(device)
pipe.scheduler = TCDScheduler.from_config(pipe.scheduler.config)

pipe.load_lora_weights(tcd_lora_id)
pipe.fuse_lora()

img_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo.png"
mask_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo_mask.png"

init_image = load_image(img_url).resize((1024, 1024))
mask_image = load_image(mask_url).resize((1024, 1024))

prompt = "a tiger sitting on a park bench"

image = pipe(
  prompt=prompt,
  image=init_image,
  mask_image=mask_image,
  num_inference_steps=8,
  guidance_scale=0,
  eta=0.3, # Eta (referred to as `gamma` in the paper) is used to control the stochasticity in every step. A value of 0.3 often yields good results.
  strength=0.99,  # make sure to use `strength` below 1.0
  generator=torch.Generator(device=device).manual_seed(0),
).images[0]

grid_image = make_image_grid([init_image, mask_image, image], rows=1, cols=3)

Versatile for Community Models

import torch
from diffusers import StableDiffusionXLPipeline
from scheduling_tcd import TCDScheduler 

device = "cuda"
base_model_id = "cagliostrolab/animagine-xl-3.0"
tcd_lora_id = "h1t/TCD-SDXL-LoRA"

pipe = StableDiffusionXLPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16, variant="fp16").to(device)
pipe.scheduler = TCDScheduler.from_config(pipe.scheduler.config)

pipe.load_lora_weights(tcd_lora_id)
pipe.fuse_lora()

prompt = "A man, clad in a meticulously tailored military uniform, stands with unwavering resolve. The uniform boasts intricate details, and his eyes gleam with determination. Strands of vibrant, windswept hair peek out from beneath the brim of his cap."

image = pipe(
    prompt=prompt,
    num_inference_steps=8,
    guidance_scale=0,
    # Eta (referred to as `gamma` in the paper) is used to control the stochasticity in every step.
    # A value of 0.3 often yields good results.
    # We recommend using a higher eta when increasing the number of inference steps.
    eta=0.3, 
    generator=torch.Generator(device=device).manual_seed(0),
).images[0]

Combine with styled LoRA

import torch
from diffusers import StableDiffusionXLPipeline
from scheduling_tcd import TCDScheduler 

device = "cuda"
base_model_id = "stabilityai/stable-diffusion-xl-base-1.0"
tcd_lora_id = "h1t/TCD-SDXL-LoRA"
styled_lora_id = "TheLastBen/Papercut_SDXL"

pipe = StableDiffusionXLPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16, variant="fp16").to(device)
pipe.scheduler = TCDScheduler.from_config(pipe.scheduler.config)

pipe.load_lora_weights(tcd_lora_id, adapter_name="tcd")
pipe.load_lora_weights(styled_lora_id, adapter_name="style")
pipe.set_adapters(["tcd", "style"], adapter_weights=[1.0, 1.0])

prompt = "papercut of a winter mountain, snow"

image = pipe(
    prompt=prompt,
    num_inference_steps=4,
    guidance_scale=0,
    # Eta (referred to as `gamma` in the paper) is used to control the stochasticity in every step.
    # A value of 0.3 often yields good results.
    # We recommend using a higher eta when increasing the number of inference steps.
    eta=0.3, 
    generator=torch.Generator(device=device).manual_seed(0),
).images[0]

Compatibility with ControlNet

Depth ControlNet

import torch
import numpy as np
from PIL import Image
from transformers import DPTFeatureExtractor, DPTForDepthEstimation
from diffusers import ControlNetModel, StableDiffusionXLControlNetPipeline
from diffusers.utils import load_image, make_image_grid
from scheduling_tcd import TCDScheduler 

device = "cuda"
depth_estimator = DPTForDepthEstimation.from_pretrained("Intel/dpt-hybrid-midas").to(device)
feature_extractor = DPTFeatureExtractor.from_pretrained("Intel/dpt-hybrid-midas")

def get_depth_map(image):
    image = feature_extractor(images=image, return_tensors="pt").pixel_values.to(device)
    with torch.no_grad(), torch.autocast(device):
        depth_map = depth_estimator(image).predicted_depth

    depth_map = torch.nn.functional.interpolate(
        depth_map.unsqueeze(1),
        size=(1024, 1024),
        mode="bicubic",
        align_corners=False,
    )
    depth_min = torch.amin(depth_map, dim=[1, 2, 3], keepdim=True)
    depth_max = torch.amax(depth_map, dim=[1, 2, 3], keepdim=True)
    depth_map = (depth_map - depth_min) / (depth_max - depth_min)
    depth_image = Image.fromarray((depth_map.squeeze().cpu().numpy() * 255).astype(np.uint8))
    return depth_image

base_model_id = "stabilityai/stable-diffusion-xl-base-1.0"
controlnet_id = "diffusers/controlnet-depth-sdxl-1.0"
tcd_lora_id = "h1t/TCD-SDXL-LoRA"

controlnet = ControlNetModel.from_pretrained(controlnet_id, torch_dtype=torch.float16).to(device)
pipe = StableDiffusionXLControlNetPipeline.from_pretrained(
    base_model_id, controlnet=controlnet, torch_dtype=torch.float16, variant="fp16"
).to(device)
pipe.scheduler = TCDScheduler.from_config(pipe.scheduler.config)

pipe.load_lora_weights(tcd_lora_id)
pipe.fuse_lora()

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

prompt = "16th century portrait of a young woman, highly detailed, sharp focus"

image = pipe(
    prompt=prompt,
    image=image,
    control_image=depth_image,
    num_inference_steps=8,
    guidance_scale=0,
    eta=0.3,
    generator=torch.Generator(device=device).manual_seed(0),
).images[0]

grid_image = make_image_grid([image, depth_image], rows=1, cols=2)

📄 License

This project is licensed under the MIT license.