Valentin Kognito 365
ValentinKognito365
·
AI & ML interests
None yet
Organizations
None yet
ValentinKognito365's activity
why lora preview images looks very different during training process when using cfg=1 and cfg=4?
2
#7 opened 2 months ago
by
adwad
Great captioning model
#5 opened 3 months ago
by
ValentinKognito365
Kohya_ss now supports the model. I'm training now
45
#3 opened 5 months ago
by
dsienra
How use this in pc ?
3
#2 opened 4 months ago
by
vihanga14
6_0 or 6_1 quant
#3 opened 4 months ago
by
ValentinKognito365
Thank you for your dedication and work !
1
#4 opened 5 months ago
by
Baekdoosan
