Super Fast Image Generation in stable diffusion using LCM LoRA
20.9 هزار بار بازدید -
11 ماه پیش
-
#lcm #stablediffusion #aiimages #a1111 #aiupdates
#lcm #stablediffusion #aiimages #a1111 #aiupdates #LoRA #art
Update: LCM sampler is available in A1111 now.
https://www.seevid.ir/fa/w/icZze060TiE Introduction and sample results
LCM Latent Consistency Models,This new LoRA Model requires only 4 to 8 sampling steps instead of the usual 25 to 50 steps, to produce complete detailed images.
https://www.seevid.ir/fa/w/icZze060TiE Downloading LCM LoRAs
where to download them from
https://www.seevid.ir/fa/w/icZze060TiE using LCM in A1111
https://www.seevid.ir/fa/w/icZze060TiE LCM Settings for CFG and Sampling steps
LCM is used at 4-8 sampling steps, and 1-2 CFG scale, can be used with Euler a, but best with LCM Sampler (manual installation can be done for this, but it is not yet official in A1111)
https://www.seevid.ir/fa/w/icZze060TiE example for using LCM with after detailer and controlnets test
https://www.seevid.ir/fa/w/icZze060TiE LCM Dreamshaper_v7 checkpoint
https://www.seevid.ir/fa/w/icZze060TiE conclusion message
Computer Specs:
Laptop: Legion 5 Pro
Processor :AMD Ryzen 7 5800H , 3201 Mhz
System RAM: 16.0 GB
Graphics GPU: NVIDIA GeForce RTX 3070 Laptop GPU 8GB
LCM LoRA doesn’t require a special plugin or installation, it can be used like a normal LoRA model, The Latent Consistency Models or LCM are trained in a different way than normal LoRA models, which allows to infer images faster.
for model downloads and more reading please check the links bellow.
Hugging face page
https://huggingface.co/latent-consistency
SD 1.5 LoRA model
https://huggingface.co/latent-consistency/lcm-lora-sdv1-5/tree/main
SDXL LoRA model
https://huggingface.co/latent-consistency/lcm-lora-sdxl/tree/main
Dreamshaper v7 LCM
https://huggingface.co/SimianLuo/LCM_Dreamshaper_v7/tree/main
More reading
https://huggingface.co/blog/lcm_lora
Note: for Hires fix, I found it to work best at higher denoising strength such as 0.7, producing very reliable results with hires steps of 5 for example when using LCM LoRA.
Note: for img2img, you should double check the actual number of steps being used, for instance, if you didnt enable an option in the User Interface settings that allows using same number of steps for img2img then 15 steps would only render 5 actual steps at 0.35 denoising strength (0.35 * 15) so you should take this into account to make sure that img2img will process images at at least 4 steps in the console window.
11 ماه پیش
در تاریخ 1402/08/22 منتشر شده
است.
20,911
بـار بازدید شده