Sign In

[Community] LoRA extract

438
19.6k
1.2k
17
Updated: Oct 5, 2024
conceptlora
Verified:
SafeTensor
Type
LoRA
Stats
156
0
Reviews
Published
Apr 17, 2023
Base Model
SD 1.4
Hash
AutoV2
B8EF562E62
Please support civitai and creators by disabling adblock

LoRA extractions is not as good as the full model but it creates a compromise between size and quality and it can be used along other LoRA model with a full model. Credit goes the creator of the checkpoint in version description

Sorry, i do not accept anymore request link(s) in my comments "Suggestions for a model extraction?"

04.0417.23 Name version is limited to 10 25 character, make it impossible to see full name... see below "List of all LoRA model".

How to use it in AUTOMATIC1111 WebUI?

  • Place the LoRA model in .\stable-diffusion-webui\models\Lora then use <lora:filename:multiplier> in the prompt.

About the comparison:

  • Picture order: First = LoRA, Second = Fine-tuned Model, Third = Base Model

  • DM = Network Dimension ( 120=lower quality/low size, 320=higher quality/higher size ) same for CD = Conv Dimension (Rank)
    [CD is optionally if you wish to use it, install the extension: https://github.com/KohakuBlueleaf/a1111-sd-webui-locon (How? Extensions -> Install from URL)]
    Obviously Version "1024CV1024" or in short "1024" use more VRAM
    I did a large comparison DM+CV here

  • sample pictures does not use vea and LoRA model with SDv1.4

  • every model from (Stable Diffusion Base Model: mostly sd-v1-4.ckpt, model base is determined by highest % affinity) was extracted using Kohya_ss (Save precision: fp16)

  • Name format: <Type><DM> <CV><Name>

List of all LoRA model:

Discussion

FlamingoDelta's Avatar
FlamingoDelta

my images are getting all distorted
I am using deliberate v2

_1_'s Avatar
_1_

Some Loras, like Perfectworld small version, fail to load. Do I need an extension for those?


RuntimeError: output with shape [128, 320, 1, 1] doesn't match the broadcast shape [128, 320, 3, 3]

Edit: I installed Locon extension just in case you mislabeled it, and lo and behold I'm getting a different error

"a1111-sd-webui-locon/scripts/main.py", line 592, in rebuild_weight if len(output_shape) == 4: UnboundLocalError: local variable 'output_shape' referenced before assignment

FlamingoDelta's Avatar
FlamingoDelta

meinamix cannot be downloaded

DarkAgent's Avatar
DarkAgent
Jason Twitch 1

Why are there 2 files for each?

What is the difference?

Hr2oFlubupruk's Avatar
Hr2oFlubupruk

Lora extracts are shit for full trained models basically, it can only contain a few concepts accurately. Why have people still not figure this out, it's in every documentation. This is bullshit and misinforming.. you won't save space on your harddrive you'll simply have shitty compressed version of full models that will have a some influence on your generations.

JaysonPhoenix's Avatar
JaysonPhoenix

can you do one of LazyMix+ v2 and maybe v1 too pls?

monkeyman007's Avatar
monkeyman007
ODOR Badge

Why is the base model for most Loras is Sd 1.4? Most people actually use SD1.5. But, anyways thanks for your hardwork.

xRikishi's Avatar
xRikishi

🤨