dealign.ai PRO
dealignai
AI & ML interests
ML Researcher - No fine tuning, no lora, no template bs, pure real ablation. - I have an obsession with making high quality red teaming LLM’s
Recent Activity
updated a model about 3 hours ago
JANGQ-AI/Ornith-1.0-397B-JANG_1L-AWQ published a model about 4 hours ago
JANGQ-AI/Ornith-1.0-397B-JANG_1L-AWQ updated a model about 5 hours ago
JANGQ-AI/Ornith-1.0-397B-JANG_1LOrganizations
Interesting work!
1
#1 opened 23 days ago
by dealignai
Fix loading in LM Studio / mlx-llm: quant map keys + vision weights
4
#25 opened about 2 months ago
by 
IsaacLiu
Gemma 4 26B (A4B) - Garbage output (korean/chinese) on MLX 0.31.2 while 31B works fine
3
#2 opened 3 months ago
by 
Z3NN001
Add MMLU-200 (93.5%), speed/memory benchmarks, fix variants table sizes (47→56 GB), expand topic tags
#1 opened about 2 months ago
by dealignai
gemma4-crack
#8 opened 3 months ago
by 
Adam666Eves
lm-studio
2
#2 opened 3 months ago
by 
hady20100
turboquant module?
1
#1 opened 3 months ago
by 
abbuu
Can you please release GGUF versions ?
7
#1 opened 3 months ago
by 
PaPaai99
Refusal rate very high on questions beyond cybersecurity
8
#4 opened 3 months ago
by 
Zayaan
Performance Test on MacBook Max – 77 tokens/s for a 1k‑token Task
1
#1 opened 3 months ago
by 
cutiedeng
JANG 3L version available in Uncensored?
4
#1 opened 3 months ago
by 
JM-account
Doesn't work for me
6
#1 opened 3 months ago
by 
x-polyglot-x
Doesn't work for me
6
#1 opened 3 months ago
by x-polyglot-x
Doesn't work for me
6
#1 opened 3 months ago
by x-polyglot-x