Crumb

RR∆S®MinoriMirari®.Prod

u/MinoriMirariRProductions@lemmy.world
7 posts · 319 comments

Recent posts

Recent comments

on Content Warning · c/perchance · 2 pts · 179d

Bro absolutely no description, and this is a repost. please🙏 dont spam the forums.

on any workaround for the negative prompts not working right now? · c/perchance · 1 pts · 180d

I dont have much news on negatives, on one my replies on 1/3 posts about it I left a note about working on new formula blocking and formatting..ect. I have plans for prebaking, and hooking them into that, of which on other 1/3 posts on negatives I keft further notes abiut Klein being a fairly technical model I believe with hints towards the prior..ect. I hope all is going well @Alllo@lemmy.world
Been awhile since I seen yoy around here, atleast any activ3 msging from you lol. 👋👌

on Where is the "negative prompt" in the ai-text-to-image-generator? · c/perchance · -1 pts · 181d

technically negatives should go in first, because negation is suppose to preload, before the rest of the prompt.. Im working on within my formula blocka advanced negation logics, at the moment Gemini and few others tipped me off that the Archaic F fire trigger/reinitialization functions will come handy for Klein, been along while since I leveraged them, an essentially because Klein runs so fast It makes data tabling sorta powerful as far as super distillation is concerned anway.. Hard to explaon why reinitialization operations are of much help here maybe because you can basically run the data into post and then re fire it ig.. although what makes Klein great for this makes it "atm" the majority of tactical formula compartmentalization decapartum phasing "all formula formatting considered to be tactical like that originally previewed and concept with TF∅X formula formatting/&series, are basically extremely touchy, if not compete broke for the most part, if just broke AF it's because Klein has the heaviest load specified sampling sub routines ever conceived" It basickly has some low level compartmentalization of its raw data tables, ontop of already having specific prior flux.1 Sampling Auto blending over blend autonomous blending routines.

on Please switch back to the old design! · c/perchance · -2 pts · 181d

If you mean Perchance OFC T2i plugin's GAi core modeling and networking,.. Its, because "Perchance COs, main dev, and even myself "Assistant Dev" can see that when it comes to an efficient model vs the prior, all changes are made to cap the cost of Ai image generation, the single most expensive system too operate here on Perchance – is the T2i plugin".. The current new model of 2026 is not fairly much diffrent from the prior Schell/Chroma/Dev mixed multimodal/Perchance customs flux.1 networking of core modeling, now on "Klein, flux.2 networking, Klein is part of the new era of super distillation models, while being technical prompting stteam‐lined". The new model, it is quite literally the most quality efficient model to date, & manages to be built to throw more resource to mid ground if being technical prompt prodded, future potential also hints at it being extremely well equip for tool building, plus leveraging of LoRa sub modeling, possibly later formalizations of sub modality and user capable loading and off loading of training data..ect. At the moment to some, I'd say few.. It may seem like complete garbage, or a literal direct downgrading of our prior customs,.. particularly because it handles as if It's nearly the same model.., It is not, and moving forward, that and users actually studying up on siht will realise "that it is actually possibly ¿A better model?". With time and good practice I'am betting the new model will prove to be "Yes better"...