Is nsfw v8 a fp8 model??
1
#45 opened 1 day ago
by
ZXWXZLIN
Feedback on the V8 model.
#44 opened 1 day ago
by
lwp513202035
Crashing with any LORA
2
#43 opened 3 days ago
by
jmigdelacruz

Can this model use the Lora trained with Wan2.2? This can keep the characters in the picture with their own Lora facial features
1
#40 opened 3 days ago
by
ELGTY

Bouncing effect
2
#39 opened 3 days ago
by
Gabby604
Fun Control Model
1
#38 opened 4 days ago
by
animega80

This model still consumes a lot of memory, How can we quantify this model?
โ
1
5
#37 opened 4 days ago
by
chfm
The v8 version seems to completely shield NSFW content, though it has slightly better quality, I still prefer the v7 version.
4
#36 opened 4 days ago
by
chfm
Itโs great to see the new version released.
#35 opened 4 days ago
by
chfm
This is a great work; looking forward to an even more exciting next version.
โค๏ธ
3
#34 opened 5 days ago
by
chfm
lots of noise in the images
4
#31 opened 7 days ago
by
skyrim
The i2v_nsfw version of v7 has some undesirable effects such as inconsistent facial features of the characters, distorted faces, and numerous noise points in the picture.
10
#30 opened 8 days ago
by
wange1002
image noise
1
#29 opened 9 days ago
by
q27916430
In version 7, LoRA often distorts the character's appearance. Please remove LoRA.
3
#28 opened 9 days ago
by
ilkolk
Is thete Wan-VideoStartEndFrames workflow plan?
๐
2
1
#27 opened 10 days ago
by
ourhome233
The V5 wan2.2-i2v-rapid-aio generate video with much worse quality than V1, could you fixed it?
1
#26 opened 11 days ago
by
richuyouyaojie
Temp first noisy frames fix
๐
2
1
#25 opened 12 days ago
by
100thape
"Please make Wan2.2_T2V_14B_VACE an all-in-one package."
1
#24 opened 12 days ago
by
ilkolk
make v6 gguf plz
1
#23 opened 12 days ago
by
xueqing12
About the Lightning model
1
#21 opened 13 days ago
by
yuyuyu012
High and Low?
3
#20 opened 14 days ago
by
sdfasdfasdfsdfsdf
Torch compile comparability:
1
#19 opened 16 days ago
by
lhucklen
FLF2V
๐
1
1
#18 opened 16 days ago
by
dk38
Is V3 supposed to be significantly slower than the first one?
5
#17 opened 17 days ago
by
AlexTheSandGrinder
The initial version gave the best 2.2 I2V results I've gotten so far, but every subsequent version melts and noises the images right from the first frames.
4
#16 opened 17 days ago
by
OlhoDeBila
v4 with new wan 2.2? lora
6
#15 opened 17 days ago
by
markasd

i hope some can create gguf
#14 opened 18 days ago
by
Kutches

I am getting this error trying to run the example json
1
#13 opened 19 days ago
by
makemoneyrp

Settings For 2
1
#12 opened 20 days ago
by
lhucklen
Sampler Path
1
#11 opened 21 days ago
by
skyrim
I'm willing to help with the FP16 version
4
#10 opened 21 days ago
by
TekeshiX

Which version of the fast LoRA is included in I2v model?
๐
2
1
#8 opened 22 days ago
by
vanditking
Please merge and generate the FP16 model to improve the accuracy. Thank you
1
#7 opened 23 days ago
by
Nurburgring
Does this model support Kijai's workflow? The workflow of Kijai can have its memory requirements optimized to a lower level.
#6 opened 23 days ago
by
wange1002
gguf plz
๐
๐
13
#5 opened 23 days ago
by
xueqing12
Loras?
๐
1
2
#4 opened 23 days ago
by
Nedo
Huge Improvement
๐
1
8
#3 opened 23 days ago
by
JohnnyAction
can u make seperate
1
#2 opened 23 days ago
by
markasd

Any result examples? [V1 examples inside, outdated]
๐
1
10
#1 opened 23 days ago
by
infoperobueno
