Low-Rank Adaptation of Large Language Models (LoRA)

Low-Rank Adaptation of Large Language Models (LoRA) is a training method that accelerates the training of large models while consuming less memory. It adds pairs of rank-decomposition weight matrices (called update matrices) to existing weights, and only trains those newly added weights. This has a couple of advantages:

• Previous pretrained weights are kept frozen so the model is not as prone to catastrophic forgetting.
• Rank-decomposition matrices have significantly fewer parameters than the original model, which means that trained LoRA weights are easily portable.
• LoRA matrices are generally added to the attention layers of the original model. 🧨 Diffusers provides the load_attn_procs() method to load the LoRA weights into a model’s attention layers. You can control the extent to which the model is adapted toward new training images via a scale parameter.
• The greater memory-efficiency allows you to run fine-tuning on consumer GPUs like the Tesla T4, RTX 3080 or even the RTX 2080 Ti! GPUs like the T4 are free and readily accessible in Kaggle or Google Colab notebooks.

cloneofsimo was the first to try out LoRA training for Stable Diffusion in the popular lora GitHub repository. 🧨 Diffusers now supports finetuning with LoRA for text-to-image generation and DreamBooth. This guide will show you how to do both.

If you’d like to store or share your model with the community, login to your Hugging Face account (create one if you don’t have one already):

huggingface-cli login

Text-to-image

Finetuning a model like Stable Diffusion, which has billions of parameters, can be slow and difficult. With LoRA, it is much easier and faster to finetune a diffusion model. It can run on hardware with as little as 11GB of GPU RAM without resorting to tricks such as 8-bit optimizers.

Training

Let’s finetune stable-diffusion-v1-5 on the Pokémon BLIP captions dataset to generate your own Pokémon.

Specify the MODEL_NAME environment variable (either a Hub model repository id or a path to the directory containing the model weights) and pass it to the pretrained_model_name_or_path argument. You’ll also need to set the DATASET_NAME environment variable to the name of the dataset you want to train on. To use your own dataset, take a look at the Create a dataset for training guide.

The OUTPUT_DIR and HUB_MODEL_ID variables are optional and specify where to save the model to on the Hub:

export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export OUTPUT_DIR="/sddata/finetune/lora/pokemon"
export HUB_MODEL_ID="pokemon-lora"
export DATASET_NAME="lambdalabs/pokemon-blip-captions"

There are some flags to be aware of before you start training:

• --push_to_hub stores the trained LoRA embeddings on the Hub.
• --report_to=wandb reports and logs the training results to your Weights & Biases dashboard (as an example, take a look at this report).
• --learning_rate=1e-04, you can afford to use a higher learning rate than you normally would with LoRA.

Now you’re ready to launch the training (you can find the full training script here). Training takes about 5 hours on a 2080 Ti GPU with 11GB of RAM, and it’ll create and save model checkpoints and the pytorch_lora_weights in your repository.

accelerate launch --mixed_precision="fp16"  train_text_to_image_lora.py \
  --pretrained_model_name_or_path=$MODEL_NAME \
  --dataset_name=$DATASET_NAME \
  --dataloader_num_workers=8 \
  --resolution=512 --center_crop --random_flip \
  --train_batch_size=1 \
  --gradient_accumulation_steps=4 \
  --max_train_steps=15000 \
  --learning_rate=1e-04 \
  --max_grad_norm=1 \
  --lr_scheduler="cosine" --lr_warmup_steps=0 \
  --output_dir=${OUTPUT_DIR} \
  --push_to_hub \
  --hub_model_id=${HUB_MODEL_ID} \
  --report_to=wandb \
  --checkpointing_steps=500 \
  --validation_prompt="A pokemon with blue eyes." \
  --seed=1337

Inference

Now you can use the model for inference by loading the base model in the StableDiffusionPipeline and then the DPMSolverMultistepScheduler:

>>> import torch
>>> from diffusers import StableDiffusionPipeline, DPMSolverMultistepScheduler

>>> model_base = "runwayml/stable-diffusion-v1-5"

>>> pipe = StableDiffusionPipeline.from_pretrained(model_base, torch_dtype=torch.float16)
>>> pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)

Load the LoRA weights from your finetuned model on top of the base model weights, and then move the pipeline to a GPU for faster inference. When you merge the LoRA weights with the frozen pretrained model weights, you can optionally adjust how much of the weights to merge with the scale parameter:

>>> pipe.unet.load_attn_procs(lora_model_path)
>>> pipe.to("cuda")
# use half the weights from the LoRA finetuned model and half the weights from the base model

>>> image = pipe(
...     "A pokemon with blue eyes.", num_inference_steps=25, guidance_scale=7.5, cross_attention_kwargs={"scale": 0.5}
... ).images[0]
# use the weights from the fully finetuned LoRA model

>>> image = pipe("A pokemon with blue eyes.", num_inference_steps=25, guidance_scale=7.5).images[0]
>>> image.save("blue_pokemon.png")

from huggingface_hub.repocard import RepoCard

lora_model_id = "sayakpaul/sd-model-finetuned-lora-t4"
card = RepoCard.load(lora_model_id)
base_model_id = card.data.to_dict()["base_model"]

pipe = StableDiffusionPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16)
...

DreamBooth

DreamBooth is a finetuning technique for personalizing a text-to-image model like Stable Diffusion to generate photorealistic images of a subject in different contexts, given a few images of the subject. However, DreamBooth is very sensitive to hyperparameters and it is easy to overfit. Some important hyperparameters to consider include those that affect the training time (learning rate, number of training steps), and inference time (number of steps, scheduler type).

Training

Let’s finetune stable-diffusion-v1-5 with DreamBooth and LoRA with some 🐶 dog images. Download and save these images to a directory. To use your own dataset, take a look at the Create a dataset for training guide.

To start, specify the MODEL_NAME environment variable (either a Hub model repository id or a path to the directory containing the model weights) and pass it to the pretrained_model_name_or_path argument. You’ll also need to set INSTANCE_DIR to the path of the directory containing the images.

The OUTPUT_DIR variables is optional and specifies where to save the model to on the Hub:

export MODEL_NAME="runwayml/stable-diffusion-v1-5"
export INSTANCE_DIR="path-to-instance-images"
export OUTPUT_DIR="path-to-save-model"

There are some flags to be aware of before you start training:

• --push_to_hub stores the trained LoRA embeddings on the Hub.
• --report_to=wandb reports and logs the training results to your Weights & Biases dashboard (as an example, take a look at this report).
• --learning_rate=1e-04, you can afford to use a higher learning rate than you normally would with LoRA.

Now you’re ready to launch the training (you can find the full training script here). The script creates and saves model checkpoints and the pytorch_lora_weights.bin file in your repository.

It’s also possible to additionally fine-tune the text encoder with LoRA. This, in most cases, leads to better results with a slight increase in the compute. To allow fine-tuning the text encoder with LoRA, specify the --train_text_encoder while launching the train_dreambooth_lora.py script.

accelerate launch train_dreambooth_lora.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --instance_data_dir=$INSTANCE_DIR \
  --output_dir=$OUTPUT_DIR \
  --instance_prompt="a photo of sks dog" \
  --resolution=512 \
  --train_batch_size=1 \
  --gradient_accumulation_steps=1 \
  --checkpointing_steps=100 \
  --learning_rate=1e-4 \
  --report_to="wandb" \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --max_train_steps=500 \
  --validation_prompt="A photo of sks dog in a bucket" \
  --validation_epochs=50 \
  --seed="0" \
  --push_to_hub

Inference

Now you can use the model for inference by loading the base model in the StableDiffusionPipeline:

>>> import torch
>>> from diffusers import StableDiffusionPipeline

>>> model_base = "runwayml/stable-diffusion-v1-5"

>>> pipe = StableDiffusionPipeline.from_pretrained(model_base, torch_dtype=torch.float16)

Load the LoRA weights from your finetuned DreamBooth model on top of the base model weights, and then move the pipeline to a GPU for faster inference. When you merge the LoRA weights with the frozen pretrained model weights, you can optionally adjust how much of the weights to merge with the scale parameter:

>>> pipe.unet.load_attn_procs(lora_model_path)
>>> pipe.to("cuda")
# use half the weights from the LoRA finetuned model and half the weights from the base model

>>> image = pipe(
...     "A picture of a sks dog in a bucket.",
...     num_inference_steps=25,
...     guidance_scale=7.5,
...     cross_attention_kwargs={"scale": 0.5},
... ).images[0]
# use the weights from the fully finetuned LoRA model

>>> image = pipe("A picture of a sks dog in a bucket.", num_inference_steps=25, guidance_scale=7.5).images[0]
>>> image.save("bucket-dog.png")

If you used --train_text_encoder during training, then use pipe.load_lora_weights() to load the LoRA weights. For example:

from huggingface_hub.repocard import RepoCard
from diffusers import StableDiffusionPipeline
import torch

lora_model_id = "sayakpaul/dreambooth-text-encoder-test"
card = RepoCard.load(lora_model_id)
base_model_id = card.data.to_dict()["base_model"]

pipe = StableDiffusionPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16)
pipe = pipe.to("cuda")
pipe.load_lora_weights(lora_model_id)
image = pipe("A picture of a sks dog in a bucket", num_inference_steps=25).images[0]

Note that the use of load_lora_weights() is preferred to load_attn_procs() for loading LoRA parameters. This is because load_lora_weights() can handle the following situations:

• LoRA parameters that don’t have separate identifiers for the UNet and the text encoder (such as "patrickvonplaten/lora_dreambooth_dog_example"). So, you can just do:

  Copied

  pipe.load_lora_weights(lora_model_path)

• LoRA parameters that have separate identifiers for the UNet and the text encoder such as: "sayakpaul/dreambooth".

Note that it is possible to provide a local directory path to load_lora_weights() as well as load_attn_procs(). To know about the supported inputs, refer to the respective docstrings.

Unloading LoRA parameters

You can call unload_lora_weights() on a pipeline to unload the LoRA parameters.

Supporting A1111 themed LoRA checkpoints from Diffusers

This support was made possible because of our amazing contributors: @takuma104 and @isidentical.

To provide seamless interoperability with A1111 to our users, we support loading A1111 formatted LoRA checkpoints using load_lora_weights() in a limited capacity. In this section, we explain how to load an A1111 formatted LoRA checkpoint from CivitAI in Diffusers and perform inference with it.

First, download a checkpoint. We’ll use this one for demonstration purposes.

wget https://civitai.com/api/download/models/15603 -O light_and_shadow.safetensors

Next, we initialize a ~DiffusionPipeline:

import torch

from diffusers import StableDiffusionPipeline, DPMSolverMultistepScheduler

pipeline = StableDiffusionPipeline.from_pretrained(
    "gsdf/Counterfeit-V2.5", torch_dtype=torch.float16, safety_checker=None
).to("cuda")
pipeline.scheduler = DPMSolverMultistepScheduler.from_config(
    pipeline.scheduler.config, use_karras_sigmas=True
)

We then load the checkpoint downloaded from CivitAI:

pipeline.load_lora_weights(".", weight_name="light_and_shadow.safetensors")

And then it’s time for running inference:

prompt = "masterpiece, best quality, 1girl, at dusk"
negative_prompt = ("(low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2), "
                   "bad composition, inaccurate eyes, extra digit, fewer digits, (extra arms:1.2), large breasts")

images = pipeline(prompt=prompt, 
    negative_prompt=negative_prompt, 
    width=512, 
    height=768, 
    num_inference_steps=15, 
    num_images_per_prompt=4,
    generator=torch.manual_seed(0)
).images

Below is a comparison between the LoRA and the non-LoRA results:

You have a similar checkpoint stored on the Hugging Face Hub, you can load it directly with load_lora_weights() like so:

lora_model_id = "sayakpaul/civitai-light-shadow-lora"
lora_filename = "light_and_shadow.safetensors"
pipeline.load_lora_weights(lora_model_id, weight_name=lora_filename)

Supporting Stable Diffusion XL LoRAs trained using the Kohya-trainer

With this PR, there should now be better support for loading Kohya-style LoRAs trained on Stable Diffusion XL (SDXL).

Here are some example checkpoints we tried out:

• SDXL 0.9:
  • https://civitai.com/models/22279?modelVersionId=118556
  • https://civitai.com/models/104515/sdxlor30costumesrevue-starlight-saijoclaudine-lora
  • https://civitai.com/models/108448/daiton-sdxl-test
  • https://filebin.net/2ntfqqnapiu9q3zx/pixelbuildings128-v1.safetensors
• SDXL 1.0:
  • https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/blob/main/sd_xl_offset_example-lora_1.0.safetensors

Here is an example of how to perform inference with these checkpoints in diffusers:

from diffusers import DiffusionPipeline
import torch 

base_model_id = "stabilityai/stable-diffusion-xl-base-0.9"
pipeline = DiffusionPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16).to("cuda")
pipeline.load_lora_weights(".", weight_name="Kamepan.safetensors")

prompt = "anime screencap, glint, drawing, best quality, light smile, shy, a full body of a girl wearing wedding dress in the middle of the forest beneath the trees, fireflies, big eyes, 2d, cute, anime girl, waifu, cel shading, magical girl, vivid colors, (outline:1.1), manga anime artstyle, masterpiece, offical wallpaper, glint <lora:kame_sdxl_v2:1>"
negative_prompt = "(deformed, bad quality, sketch, depth of field, blurry:1.1), grainy, bad anatomy, bad perspective, old, ugly, realistic, cartoon, disney, bad propotions"
generator = torch.manual_seed(2947883060)
num_inference_steps = 30
guidance_scale = 7

image = pipeline(
    prompt=prompt, negative_prompt=negative_prompt, num_inference_steps=num_inference_steps,
    generator=generator, guidance_scale=guidance_scale
).images[0]
image.save("Kamepan.png")

Kamepan.safetensors comes from https://civitai.com/models/22279?modelVersionId=118556 .

If you notice carefully, the inference UX is exactly identical to what we presented in the sections above.

Thanks to @isidentical for helping us on integrating this feature.

Known limitations specific to the Kohya-styled LoRAs

• SDXL LoRAs that have both the text encoders are currently leading to weird results. We’re actively investigating the issue.
• When images don’t looks similar to other UIs such ComfyUI, it can be beacause of multiple reasons as explained here.