Fi am testing flux fast training of civitAI to compare
F
UActually I tried to merge this features with flux so that it would create the images then convert it to vector with the specified features (color, stroke, style, etc) but I couldn't. I'm not that good at coding I think.
Ftry newest chatgpt
Fi use it on poe
UWill do
Fcivit ai fast flux is really decent at distant shots
Fbut at closer shots you see its quality not good
Ais kohya actually any different than onetrainer?
Fstable diffusion mods are out of control
F
S@Dr. Furkan Gözükara I noticed all your generations have very clear backgrounds, how is that possible? Mine always come out with a shallow depth of field which is soft and blurry background
Fi just replied back on youtube :d
Fthe LoRA training making that impact. As you do train more epochs, that annoying "shallow depth of field which is soft and often blurry" effect is gone. But more training also causes overfitting so you have to have a very good config + decide which checkpoint to use for balance both
Mdidnt one of the mods say its fine earlier ??
Fthere is 1 mod i thin he is responsible
Fso annoying
Mi think thats one of your biggests posts too
Mlot of comments seemed happy
Fyep huge upvote
Fit was one of the most viral posts
SOh no, damn! This was the flux post with those many awesome images ?
Ftime to do some fine tuning
Mjust got word that sd3 8b will be out soon, no distilation and easier to train and finetune
Flicence?
Mi think same as the the updated sd3 medium one with more clarity
Mlykon mentioned
Mlooks like it can be better than flux
Fnice
M@linaqruf_ @oron1208 Wait for 8b. It's basically Flux without distillation and heavy hands dpo. This should make it easy to finetune (and dpo).
We're also trying a new scaling down mechanism for mmdit, the new 2b is gonna work much better.
We're also trying a new scaling down mechanism for mmdit, the new 2b is gonna work much better.
9/14/24, 11:30 AM
Ffull fine tuning
Fnice
Fit is also fp16 atm
Fnot even fp8 it will be lower
Myou can try training on a a100 to see if results are good first
Fnot needed
Fit works perfect on A6000
Ffused back pass reduced VRAM usage by far
Falso there are block swapping
Fso it will work on 24gb too
Fbut fp8 didnt make any diff
Mbut very slow though right?
Ffor some reason
Fsame speed as lora
Mwith block swapping?
Fno without
Fwith block it will be slower i expect
F