yes that is because of the distilled guidance right?
yes that is because of the distilled guidance right?
Cand that makes it impossible to do anything with other than what it is
Cno. that just makes it worse.
Cno, that's becuase of other things that were done before that
Dyou always get something coherent
Csure, because it uses the t5xxl encoder nd the clip_l encoder. so you're going to get coherant. but give it a word like Umber and you should get mostly the color. the dictionary defintion for umber is "a natural pigment resembling but darker than ocher, normally dark yellowish-brown in color ( raw umber ) or dark brown when roasted ( burnt umber ).
2.
a brownish-gray moth with coloring that resembles tree bark."
2.
a brownish-gray moth with coloring that resembles tree bark."
Cthat's not what you get
Cyou should get random subjects, since umber isn't a subject - all over the map
Cyou don't get that
Cyou get very specific subjects as that's what it draws when you don't tell it a subject
Dyou put something random and and get a dog for example
DI know what you said
Cif i tell you this word "Unboundedness" - what would you tell me it looks like?
Dhehe
Dno clue
Dthe model probably returns a cute anime cat
Dhehe
Cit should return random images with random subjects. because it doesn't have a subject so it should just be pulling up data from all over latent space
Cif you put that word into sd 1.5, or sdxl, or midjourney - you'll get interesting results
Cput it in flux dev and see what you get
Dthanks for your detailed explanation, let's see how this ends up
Cbased on the blowup that he had a few weeks back, he'll refuse to admit he cna't do what he wants. he'll continue on till he gets 'something' and then boast about it
Cand it'll return some sorts of results, and people will believe he undistilled flux because they don't know any better. and blackforest will sit there and laugh at him
DI know that is imposible to convert a distilled model to its original state, the information that was lost in the process is lost forever
Cthat's true too. you spend the money, you no longer have the money
Cwe get requests along those lines for Photoshop all the time "hi, i have this photo of this car, can you make it so i can read the license' - and the plate is just a black and white blur. 'um, no. this isn't camera raw footage, it's a jpg"
Dwhat could be a success is to make a model that is more flexible and more trainable, will not be the black forest pro model, will be other model that is more trainable
Dthat if its possible
Cto do that, robin will need to scrap flux and start over. and that was at least a million, possibly more, to train. that's a huge hit now that he's nto at stability.ai
Cso he's not likely to do that
Cwhat he's most likely to do is continue to develop Pro - it's api only, it's making him money, he can sell it to top movie studios and commercial businesses
Donly fixing the human class will be a success
Cand IF they want special loras, he can deal with that for them himself
Ctrust me on this, okay? he tried. that's why it's so overfit for women
Cor, well, maybe it's just heavily biased for women. same effect however. here's the phrase "Pastel pencil". two women, a dog, and a rose
Cno landscapes, or anything else. 3 out of 4 are women and dogs
Cwith an anime look to one of the women
Cand i jsut generated that
PAhh nice! Thankls for the tip 
Cand the thing that using only "Pastel pencil" SHOULD have returned is - 4 images of actual pencils
Cyou don't see pencils at all
Dyou can train with a very diverse people all genders all races all ages 3000 pictures at least for many epochs and may be end up more flexible and training on top produces lees bleeding, that will be a success for me, idk
Cthat would require flux to be trainable. it isn't.
Cnot even robin can train it. he'd have to train the base model, and that's expensive. VERY expensive. hundreds of dollars at the minimum, but more like thousands of dollars if not more
Ci was seriou when i said that training the base model was at least a million dollars
Cand for him to train new data into it, that's not making a lora, that's running the training on the base model on super clusters of GPUS for quite a while.
Dyes but training not the entire model, just the human class
Dthat will be cheaper
Cthat requires making a lora. that's what loras are for. and that can not be done
Che would have to train the base model and update the weights in it
Dyou can do a dreambooth