Masterjp123 commited on
Commit
062f359
1 Parent(s): 0d78f67

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -4
README.md CHANGED
@@ -14,16 +14,26 @@ Well Someone who is way more skilled at LLM Training and Lora making, should mak
14
  Then Well Recreate MythosMax, since it brought a new Era of RP LLMs, if it was recreated in LLama3 once people figure out how to fully Crack LLama 3 to be fully uncensored,
15
  it will make it so the models will be better, since Any Model that used MythosMax had better performance.
16
 
 
 
17
  ## 2: TTS (Diffusion Models)
18
 
19
  SD Models:
20
 
21
  With the current state of SD3 not getting released, my faith in SAI Is waning, So Time to look for another company, Or person.
22
 
 
 
 
 
 
 
 
 
23
  Diffusion Models:
24
 
25
- For Diffusion models, if we want a better model than SD1.5 We'll need a replacement of it which is the 600-800 Milion parameter range, because currently SD1.5 is good... But there are way better, the only problem:
26
- THEY ARE OVER 1B AND FOR DIFFUSION MODELS THEY ALREADY ARE HARD TO RUN, Because A LLama 2 model gives me about 20t/S While Sd1.5 models give me only 1.4it/s
27
 
28
- ## Licence for My Ideas and technices that might be here one day:
29
- Well Use'em how ever you'd like, just like put my name somewhere.
 
 
14
  Then Well Recreate MythosMax, since it brought a new Era of RP LLMs, if it was recreated in LLama3 once people figure out how to fully Crack LLama 3 to be fully uncensored,
15
  it will make it so the models will be better, since Any Model that used MythosMax had better performance.
16
 
17
+ NOTE: Stheno 3.2 fixed all of the problems now it just needs to be finetuned more.
18
+
19
  ## 2: TTS (Diffusion Models)
20
 
21
  SD Models:
22
 
23
  With the current state of SD3 not getting released, my faith in SAI Is waning, So Time to look for another company, Or person.
24
 
25
+ UPDATE On Stability Ai:
26
+ Damm They really just shot themselves in the foot huh...
27
+
28
+ <iframe src="https://giphy.com/embed/WtBDAH97eXAmQ" width="480" height="341" style="" frameBorder="0" class="giphy-embed" allowFullScreen></iframe><p><a href="https://giphy.com/gifs/postal-apocalypse-WtBDAH97eXAmQ">via GIPHY</a></p>
29
+
30
+ Damm it Stability AI, You guys really did crap the bed after Emad left, after all if SDXL licensing wasn't bad enough, SD3 licicing is more crap than what the internet is giving them right now.
31
+ Anyways the only way Stability AI can keep their reputation with the open source community is if they fix their licicing and also maybe release the 800 Million Parameter version of SD3.
32
+
33
  Diffusion Models:
34
 
35
+ For Diffusion models the best way to basically dig Stability AI’s grave which might be the only way to get good models with better licensing, is to release a model with the upsides of a DIT model with the size around Sd1.5 which will make the ULTIMATE MODEL: Patent NOT pending.
 
36
 
37
+ ## License for My Ideas and techniques that might be here one day:
38
+ Well Use'em how ever you'd like, just like put my name somewhere and make the thing you make open source since that might be nice...
39
+ Also probably don't take my advice lol, After all I can't even create an Ai model I only merge them :(