LCM-LoRA Weights - Stable Diffusion Acceleration Module - LCM for SD 1.5
Highlighted images
Recommended Prompts
beautiful scenery nature glass bottle landscape, purple galaxy bottle
Recommended Negative Prompts
text, watermark
Recommended Parameters
samplers
steps
cfg
Tips
Use LCM-LoRA alongside any SDXL or SD 1.5 model for acceleration.
Set CFG value approximately to 1.5 and steps to 3 for optimal speed and quality.
On high-end GPUs like Nvidia RTX 4090, image generation happens nearly instantaneously.
Basic LCM ComfyUI workflow and training images are available in a zip format.
Creator Sponsors
Read more about LCM-LoRA acceleration module in the LCM-LoRA Technical Report.
Download official models here: LCM-LoRA SDXL and LCM-LoRA SD 1.5.
LCM-LoRA - Acceleration Module!
Tested with ComfyUI, although I hear it's working with Auto1111 now!
Step 1) Download LoRA
Step 2) Add LoRA alongside any SDXL Model (or 1.5 model, if using the SD 1.5 version)
Step 3) Set CFG to ~1.5 and Steps to 3
Step 4) Generate images in ~<1 second (instantaneously on a 4090)
Basic LCM Comfy workflow attached as "Training Images" in zip format.
Read more about the concept here;
https://github.com/luosiallen/latent-consistency-model/blob/main/LCM-LoRA%20Technical%20Report/LCM-LoRA-Technical-Report.pdf
Original HF link - SDXL; https://huggingface.co/latent-consistency/lcm-lora-sdxl/tree/main
Original HF link - SD 1.5; https://huggingface.co/latent-consistency/lcm-lora-sdv1-5
Model Details
Discussion
Please log in to leave a comment.