Sign In

ThisIsntAI | Beyond Of Your Imagination

109
836
248
44
Updated: Feb 13, 2025
style
Verified:
SafeTensor
Type
LoRA
Stats
83
28
Reviews
Published
Feb 11, 2025
Base Model
SDXL 1.0
Trigger Words
photo_background
Hash
AutoV2
9130FDC76E
default creator card background decoration
Y_X's Avatar
Y_X

-------------𝔹𝕖π•ͺπ• π•Ÿπ•• 𝕆𝕗 𝕐𝕠𝕦𝕣 π•€π•žπ•’π•˜π•šπ•Ÿπ•’π•₯π•šπ• π•Ÿ----------

NOTE: ryosios text encoder (NoobAI fault, not my lora) is too heavy, try to lower down weight

2/6/2025: Looks like big lora size is not a great idea. Loss value is useless..?

2/7/2025: Loss value is useful... TIA is very close to higher level AND smaller lora file sizeπŸ₯³πŸ₯³πŸ₯³

2/8/2025: paper thingy

Self-rectifying Diffusion Sampling with Perturbed-Attention Guidance |  SpringerLink

2/10/2025: CLIP Improvement (Cancelled, Fail)

2/11/2025: Due to CLIP Limitations, ThisIsntAI cannot go further.

This is where my broken lora or the good ones are posted here.

This is settings for Euler Fan.:

Euler A
Align Your Steps GITS or Align Your Steps
Sampling steps: more than 20
Mahiro CFG for reForge or comfyui Enabled  
CFG Scale: 4-5
No Negative

This is settings Another:

Restart 
Align Your Steps GITS
Sampling steps: 10-20
Mahiro CFG for reForge or comfyui Enabled  
CFG Scale: 4
No Negative

This is settings Best:

https://github.com/pamparamm/sd-perturbed-attention

Restart 
Align Your Steps GITS
Sampling steps: 10 or 20
Mahiro CFG for reForge or comfyui Enabled  
CFG Scale: 1-2
No Negative