all
/
g
/
s
/
fur
/
art
/
ai
/
3dcg
/
f
/
rp
/
req
/
p2p
/
dis
/
lit
/
kaki
chat
GUROchan!
This list is WIP...
1
2
3
So use the classic menu
Link 4
Link 5
Link 6
Link 7
List 3
Link 8
Link 9
[
Settings
]
[
Home
] [
Catalog
] [
Search
] [
Thread list
] [
Stats
] [
Reports
] [
Watcher
] [
PMs
] [
Admin
]
Artificial Intelligence
[
Return
]
Report a post
Preview
galathien
◆0AdiozdanE
[
PM
]
2026/02/14
(Sat)
22:04:26
No.
16053
[
Report
]
▶
>>16052
> What kind of Vida card were you using for Wan? I'm on Wan 2.2 right now. I'm outputting 1920x1072, 27 frames and it's taking 10 min to do only 3 seconds. If I drop the resolution down to 480, it's a chunky mess. There must be some post processing being done, cus your videos are crisp.
>
> Right as I finished the above paragraph, I got 5 seconds of 720p out of Hunyuan 1.5. It took 20 mins. Grok processing must be astronimical.
None of this Grok stuff can be done locally using ComfyUI or the like. It's all being done on commercial servers with way more power than you're ever likely to have on a home PC. And clearly, even though there are content filters put in the way (that some people are managing to evade for now), the basic Grok model must have been trained on a lot of violence and gore, imagery you'll have a damned hard time ever getting out of Wan 2.2, at least not without specialized LoRAs you might very well have to train yourself.
(If there are some good gore LoRAs out there for Wan 2.2 that I haven't been able to find, I'd love to hear about them.)
Post number
No.
16053
Board
Artificial Intelligence
Reason
Optional. Describe what's wrong with it.
Style: