someone tested T5 training??
someone tested T5 training??
Fwhat a stupid app wechat is. it wants me to someone else invite me
Fnot yet
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=40)
Dhopefully u will get the sms through
Dare u back in the city now?
Dthanks I wonder if it can fix the bleeding problem 
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=16)
Fye back in city. sms not arriving
Fyes may fix
Fbut i am more hopeful of fine tuning
Di think they are overloaded
Di had to refresh there, was fine yesterday 
F:/
Fman we need open source this
XAny idea when Kohya gui will be back up or how I can do flux Lora training?
Fit is working
Fuse v30 zip file
D
Gwhat Ai video is this? open source?
Jdataset was qt good consisting of 30+ images, is there are settings to force generated images not to adjust the size of object.
Jist hailuoai.com and its really fun 
NCompared to Joy Caption, PromptGen can capture a character's facial expression, camera angle, etc.
Sguys, my lora training is almost finished, but I accidentally closed web browser with Kohya SS, what should I do?
NIf you trained on your pc, look in the name lora folder in the model folder
D
Fit will continue
Fi am gonna test it lets see if worthy to move
D@Dr. Furkan Gözükara I would love to see a tutorial and optimizations on how to train a Lora for Schnell since it's actually open source.
I get that Dev is better but I would rather be able to use my models however I want to.
I get that Dev is better but I would rather be able to use my models however I want to.
MI got this error too
Mis this about the fastapi bug or not related?
DJust use dev and say you used schnell 
or just say I used Flux1 and leave everything else to interpretation
or just say I used Flux1 and leave everything else to interpretation DHow do we get the Pro also?
SI can't add flux dev in StableSwarmUI, 1st of all the flux downloader doesn't work if the folder name is not SwarmUI, so I had to rename it.
Safter renaming it downloads ok, but doesn't showup in swarmui -> models
Susing the FLUX_models_Auto_Downloaders_v6.zip, it wants folder to be named SwarmUI and not StableSwarmUI
S16:13:14.354 [Warning] ComfyUI-0 on port 7821 stderr: !!! Exception during processing !!! ERROR: Could not detect model type of: C:\software\swarmui\StableSwarmUI\Models\Stable-Diffusion\flux1-dev.safetensors
DMove it to the unet folder
DAnd refresh
SI think it was there when it downloaded, but it doesn't work. When I run swarmUI it shows popup saying Error, model not found. And in the models tab there are only 2 models: SDXL and SD3 Medium
DThat's strange. You sure the download did not break or something?
Syes, will be redownloading now
SI just wanted to fly like superman
Dthe new SD models have been released but are not open source, I think flux is better
Fi tested it
Fit is so bad
FReddit
Explore this post and more from the SECourses community
Ffast api bug
Finaccurate folder
Fput into unet
DI'm curious. How much does the Pro cost?