cheers Doc i will give it a wirl, ill wait for the lora combiner
cheers Doc i will give it a wirl, ill wait for the lora combiner
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=40)
Dwould u suggest a fresh reinstall or an update.bat jobby
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=16)
F
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
Xlike which software? for upscale?
AWan is not there yet
Xi do notice that wan has a bit better charterer consistency then kling generations
A1hour for ai videos that require at least 5, tries to get results = 5 hours
Aif it gets to 10 minutes then we good
JFP8 not working on v31 after a fresh install
Fjust tested on rtx 5090 and 3090 ti
Fboth works
F1.3b
F
Fi recommend fresh install and move models
JI downloaded the models the first time and have never downloaded them again. I have copied them from one version to another. Can this cause problems?
Fmodels didnt change
Fi plan to add single model file download as well and if works gguf
JWAN 2.1 1.3B (Text/Video-to-Video) FP8 works.
J1.3B FP8 with teacache. Works
Jtesting WAN 2.1 14B Text-to-Video FP8
Jworks too
JWAN 2.1 14B Image-to-Video 480P is the only model that doesn't work with FP8 (1 week ago was working fine)
JI didn't test WAN 2.1 14B Image-to-Video 720P because doesn't work on my PC
Flet me test
Fdidnt work for me too :d
Flet me fix
JIt was working fine 1 week ago
Xi download my safetensor files into my model/diffusion_models file and still cant find them when i open up swarm_ui
Fye them broken not me :d
Fclick refresh
Xhmm maybe i forgot to delete the download text from the link
F\SwarmUI\Models\diffusion_models
Jcan be fixed? or we need to wait?
XRenamed one of the safetensor files inside of \SwarmUI\Models\diffusion_models and delted the "?download" and it worked
Fwait testing
Fok working gonna update now. i hope nothing else broken :d
Ftesting
Ffixed
Fplease run update file
Fit should fix
Jthanks!!
JI'm ending a video, I'll update when it ends
Fsure
JSame prompt but with 14B model (FP8)
Fwhat a difference
Fi hope nothing broken please test :d
1Hey, did I read correctly that with the new Flux Kohya training I can use the 48Gb preset with a 5090 ok?
JWAN 2.1 14B Image-to-Video 480P FP8 working fine now. Thanks!!
