--- license: creativeml-openrail-m language: - en pipeline_tag: text-to-image tags: - art library_name: diffusers --- # Overview📃✏️ This is a Diffusers-compatible version of [Yiffymix v51 by chilon249](https://civitai.com/models/3671?modelVersionId=658237). See the original page for more information. Keep in mind that this is [SDXL-Lightning](https://huggingface.co/ByteDance/SDXL-Lightning) checkpoint model, so using fewer steps (around 12 to 25) and low guidance scale (around 4 to 6) is recommended for the best result. It's also recommended to use clip skip of 2. This repository uses DPM++ 2M Karras as its sampler (Diffusers only). # Diffusers Installation🧨 ### Dependencies Installation📁 First, you'll need to install few dependencies. This is a one-time operation, you only need to run the code once. ```py !pip install -q diffusers transformers accelerate ``` ### Model Installation💿 After the installation, you can run SDXL with this repository using the code below: ```py from diffusers import StableDiffusionXLPipeline import torch model = "IDK-ab0ut/Yiffymix_v51-XL" pipeline = StableDiffusionXLPipeline.from_pretrained( model, torch_dtype=torch.float16).to("cuda") prompt = "a cat, detailed background, dynamic lighting" negative_prompt = "low resolution, bad quality, deformed" steps = 25 guidance_scale = 4 image = pipeline(prompt=prompt, negative_prompt=negative_prompt, num_inference_steps=steps, guidance_scale=guidance_scale, clip_skip=2).images[0] image ``` Feel free to edit the image's configuration with your desire. # Scheduler's Customization⚙️ ㅤㅤㅤㅤ🧨For Diffusers🧨 You can see all available schedulers [here](https://huggingface.co/docs/diffusers/v0.11.0/en/api/schedulers/overview). To use scheduler other than DPM++ 2M Karras for this repository, make sure to import the corresponding pipeline for the scheduler you want to use. For example, we want to use Euler. First, import [EulerDiscreteScheduler](https://huggingface.co/docs/diffusers/v0.29.2/en/api/schedulers/euler#diffusers.EulerDiscreteScheduler) from Diffusers by adding this line of code. ```py from diffusers import StableDiffusionXLPipeline, EulerDiscreteScheduler ``` Next step is to load the scheduler. ```py model = "IDK-ab0ut/Yiffymix_v51" euler = EulerDiscreteScheduler.from_pretrained( model, subfolder="scheduler") pipeline = StableDiffusionXLPipeline.from_pretrained( model, scheduler=euler, torch.dtype=torch.float16 ).to("cuda") ``` Now you can generate any images using the scheduler you want. Another example is using DPM++ 2M SDE Karras. We want to import [DPMSolverMultistepScheduler](https://huggingface.co/docs/diffusers/v0.29.2/api/schedulers/multistep_dpm_solver) from Diffusers first. ```py from diffusers import StableDiffusionXLPipeline, DPMSolverMultistepScheduler ``` Next, load the scheduler into the model. ```py model = "IDK-ab0ut/Yiffymix_v51" dpmsolver = DPMSolverMultistepScheduler.from_pretrained( model, subfolder="scheduler", use_karras_sigmas=True, algorithm_type="sde-dpmsolver++").to("cuda") # 'use_karras_sigmas' is called to make the scheduler # use Karras sigmas during sampling. pipeline = StableDiffusionXLPipeline.from_pretrained( model, scheduler=dpmsolver, torch.dtype=torch.float16, ).to("cuda") ``` ## Variational Autoencoder (VAE) Installation🖼 There are two ways to get [Variational Autoencoder (VAE)](https://huggingface.co/learn/computer-vision-course/en/unit5/generative-models/variational_autoencoders) file into the model. The first one is to download the file manually and the second one is to remotely download the file using code. In this repository, I'll explain the method of using code as the efficient way. First step is to download the VAE file. You can download the file manually or remotely, but I recommend you to use the remote one. Usually, VAE files are in .safetensors format. There are two websites you can visit to download VAE. Those are HuggingFace and [CivitAI](civitai.com). ### From HuggingFace😊 This method is pretty straightforward. Pick any VAE's repository you like. Then, navigate to "Files" and the VAE's file. Make sure to click the file. Click the "Copy Download Link" for the file, you'll need this. Next step is to load [AutoencoderKL](https://huggingface.co/docs/diffusers/en/api/models/autoencoderkl) pipeline into the code. ```py from diffusers import StableDiffusionXLPipeline, AutoencoderKL ``` Finally, load the VAE file into [AutoencoderKL](https://huggingface.co/docs/diffusers/en/api/models/autoencoderkl). ```py link = "your vae's link" model = "IDK-ab0ut/Yiffymix_v51" vae = AutoencoderKL.from_single_file(link).to("cuda") pipeline = StableDiffusionXLPipeline.from_pretrained( model, vae=vae).to("cuda") ``` If you're using FP16 for the model, it's essential to also use FP16 for the VAE. ```py link = "your vae's link" model = "IDK-ab0ut/Yiffymix_v51" vae = AutoencoderKL.from_single_file( link, torch_dtype=torch.float16).to("cuda") pipeline = StableDiffusionXLPipeline.from_pretrained( model, torch_dtype=torch.float16, vae=vae).to("cuda") ``` For manual download, just fill the `link` variable or any string variables you use to load the VAE file with path directory of the .safetensors. ##### Troubleshooting🔧 In case if you're experiencing `HTTP404` error because the program can't resolve your link, here's a simple fix. First, download [huggingface_hub](https://huggingface.co/docs/huggingface_hub/en/index) using `pip`. ```py !pip install --upgrade huggingface_hub ``` Import [hf_hub_download()](https://huggingface.co/docs/huggingface_hub/en/guides/download) from [huggingface_hub](https://huggingface.co/docs/huggingface_hub/en/index). ```py from huggingface_hub import hf_hub_download ``` Next, instead of direct link to the file, you want to use the repository ID. ```py repo = "username/model" file = "the vae's file.safetensors" vae = AutoencoderKL.from_single_file( hf_hub_download(repo_id=repo, filename=file)).to("cuda") # use 'torch_dtype=torch.float16' for FP16. # add 'subfolder="folder_name"' argument if the VAE is in specific folder. ``` You can use [hf_hub_download()](https://huggingface.co/docs/huggingface_hub/en/guides/download) from [huggingface_hub](https://huggingface.co/docs/huggingface_hub/en/index) casually without the need to check if previous method returns `HTTP404` error. ### From CivitAI🇨 It's trickier if the VAE is in [CivitAI](civitai.com), because you can't use `from_single_file()` method. It only works for files inside HuggingFace and local files only. You can upload the VAE from there into HuggingFace, but you must comply with the model's license before continuing. To solve this issue, you may use `wget` or `curl` command to get the file from outside HuggingFace. Before downloading, to organize the VAE file you want to use and download, change the directory to save the downloaded model with `cd`. Use `-O` option before specifying the file's link and name. It's the same thing for both `wget` and `curl`. ```py # For 'wget' !cd ; wget -O [filename.safetensors] # For 'curl' !cd ; curl -O [filename.safetensors] # Use only one of them. Replace "filename" with any # name you want. If you run the code in Command Prompt or # Windows Shell, you don't need the exclamation mark (!). ``` Since the file is now in your local directory, you can finally use `from_single_file()` method normally. Make sure to input the correct path for your VAE file. Load the VAE file into [AutoencoderKL](https://huggingface.co/docs/diffusers/en/api/models/autoencoderkl). ```py path = "path to VAE" # Ends with .safetensors file format. model = "IDK-ab0ut/Yiffymix_v51" vae = AutoencoderKL.from_single_file(path).to("cuda") pipeline = StableDiffusionXLPipeline.from_pretrained( model, vae=vae).to("cuda") # Use 'torch_dtype=torch.float16' for both # AutoencoderKL and SDXL pipeline for FP16. ``` Now you have it, loaded VAE from [CivitAI](civitai.com). # That's all for this repository. Thank you for reading my silly note. Have a nice day! #### Any helps or suggestions will be appreciated. Thank you!