LCM-LoRA版本v1.0 (ID: 223551)

LCM-LoRA - Acceleration Module

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

描述:

This is a Lora- Yet it affects the whole generation.
Stable Difussion / ComfyUI / Runpod usable
Ideal Settings for Fast Generation: there's no best sampler as it's a different style.

Combinable with other well-trained LoRas

Sampler: LCM (ComfyUI only, as of now, November 15, 2023)

Steps: 4
CFG: 2
Scheduler: Karras (For ComfyUI)

Sampler: DPM SDE ++ Karras

Steps: 4
CFG: 2
Scheduler: Karras (For ComfyUI)

Sampler: Euler A
Steps: 4
CFG: 1.3 - 2 (Alters style, try what you prefer)

Scheduler: Normal (For ComfyUI)

训练词语: