Some of the WAN models are trained with the "man" class token instead of the "woman" class token.
#11 opened about 14 hours ago
by
Pillower
greatly appreciate your hard work
2
#8 opened 23 days ago
by
whanajira
Using with Wan 2.2
1
#7 opened about 1 month ago
by
Hellomattieo
What prompt we can use to generate consistent character
1
#6 opened about 1 month ago
by
yesidevelop
We need more got chars!
1
#5 opened about 2 months ago
by
yui7854
What are your WAN Lora Training and Dataset Captioning Settings?
2
#4 opened 2 months ago
by
thesigma40323