Koitenshin
Koitenshin
AI & ML interests
I like breaking things and playing the mad scientist. Did you know we still haven't completely untapped the full potential of SD 1.5 yet?
Recent Activity
new activity 10 days ago
mistralai/Ministral-3-14B-Instruct-2512:The model suffers from ADHD new activity 15 days ago
deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:What about uncensored/abliterated version? new activity about 1 month ago
google/gemma-4-E4B-it:Image Dimension Test ResultsOrganizations
None yet
The model suffers from ADHD
👍 1
7
#18 opened 5 months ago
by 
Candala
What about uncensored/abliterated version?
❤️ 2
9
#2 opened 29 days ago
by 
tima2431
Image Dimension Test Results
1
#27 opened about 1 month ago
by Koitenshin
An Improvement, But Q3 30b Still Has Very Little General Knowledge
👍❤️ 3
11
#2 opened 10 months ago
by 
phil111
Any news on the awesome new benchmarks?
🚀👍 3
10
#373 opened 9 months ago
by 
VOIDER
Not Uncensored
13
#1 opened almost 2 years ago
by 
SicariusSicariiStuff
Possible Improvements / Future Options
👍 1
#1 opened 9 months ago
by Koitenshin
Q4_K_M is Very Unstable
#1 opened 9 months ago
by Koitenshin
Breaks once it hits Context Limit
2
#1 opened 9 months ago
by Koitenshin
Let there be light !
13
#1 opened 9 months ago
by 
phoebdroid
Comparison?
1
#5 opened 9 months ago
by 
yukiarimo
Specialized LLMs are exactly what is needed...
#1 opened 9 months ago
by Koitenshin
Hermes-Coder
❤️ 1
2
#1 opened 9 months ago
by 
Trilogix1
Model is appropriately named
2
#1 opened 9 months ago
by Koitenshin
how to make determinstic output?
4
#23 opened 10 months ago
by 
junma
Godly, but only with a Jailbreak.
#4 opened 9 months ago
by Koitenshin
Is this model vision capable?
2
#3 opened about 1 year ago
by 
lkraven
My comment
👍 1
2
#2 opened about 1 year ago
by 
Emalog
Once it admits it is wrong, it breaks
2
#7 opened 10 months ago
by Koitenshin
Content Policy Obsessed Model at Expense of Intelligence and Context Following
👍 19
31
#20 opened 10 months ago
by 
jalder1989