🔄 We deserve better quants!

Niki

NikiKrutan

·

https://t.me/nikikrutan

NikiKrutan

AI & ML interests

Making better quants

Recent Activity

new activity about 1 hour ago

OrionLLM/GRM-2.6-Plus-0628:Some thoughts

commentedon an article about 14 hours ago

DAVIDAU DELETED MY POST AND BANNED ME FOR ASKING ABOUT BENCHMARKS — HERE IS THE FULL ANALYSIS (9B + 27B + 390 MODELS)

commentedon an article about 14 hours ago

DAVIDAU DELETED MY POST AND BANNED ME FOR ASKING ABOUT BENCHMARKS — HERE IS THE FULL ANALYSIS (9B + 27B + 390 MODELS)

View all activity

Organizations

None yet

New activity in OrionLLM/GRM-2.6-Plus-0628 about 1 hour ago

Some thoughts

#5 opened about 22 hours ago by

commented on DAVIDAU DELETED MY POST AND BANNED ME FOR ASKING ABOUT BENCHMARKS — HERE IS THE FULL ANALYSIS (9B + 27B + 390 MODELS) about 14 hours ago

40B model is strange. Less arc. Why better then?
nightmedia already has 740 arc for 27B sitting silently.

commented on DAVIDAU DELETED MY POST AND BANNED ME FOR ASKING ABOUT BENCHMARKS — HERE IS THE FULL ANALYSIS (9B + 27B + 390 MODELS) about 14 hours ago

I've tested DavidAU's quants of 711 while making my own quants of that model. Output in 16-bit is just waste of model size. "LOW" quants are good on wiki though.
Full evidence here: https://huggingface.co/NikiKrutan/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP-GGUF

upvoted an article about 14 hours ago

Article

DAVIDAU DELETED MY POST AND BANNED ME FOR ASKING ABOUT BENCHMARKS — HERE IS THE FULL ANALYSIS (9B + 27B + 390 MODELS)

DedeProGames

•

about 17 hours ago

• 8

updated a model about 23 hours ago

NikiKrutan/GRM-2.6-Plus-0628-MTP-GGUF

Image-Text-to-Text • 27B • Updated about 23 hours ago • 2.87k • 1

updated a model 1 day ago

NikiKrutan/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP-GGUF

Image-Text-to-Text • 27B • Updated 1 day ago • 5

liked 2 models 1 day ago

OrionLLM/GRM-3.2-Sky

Image-Text-to-Text • 665k • Updated 10 days ago • 843 • 35

OrionLLM/LRM-3.2

Text Generation • 27B • Updated about 23 hours ago • 8

New activity in DavidAU/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-NEO-MAX-MTP-GGUF 1 day ago

Somewhat better quants here

#79 opened 1 day ago by

published a model 2 days ago

NikiKrutan/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP-GGUF

Image-Text-to-Text • 27B • Updated 1 day ago • 5

New activity in OrionLLM/GRM-2.6-Plus-0628 5 days ago

GGUF quants

#4 opened 5 days ago by

liked a model 6 days ago

nightmedia/Qwen3.6-27B-Jormungandr

Image-Text-to-Text • 28B • Updated 8 days ago • 49 • 3

updated a model 6 days ago

NikiKrutan/Qwen3.5-0.8B-MTP-GGUF

Image-Text-to-Text • 0.8B • Updated 6 days ago • 5.88k • 1

published a model 6 days ago

NikiKrutan/GRM-2.6-Plus-0628-MTP-GGUF

Image-Text-to-Text • 27B • Updated about 23 hours ago • 2.87k • 1

New activity in Moore2877/Qwen-Fixed-Chat-Templates-llamacpp 7 days ago

<tool_call> inside reasoning

#3 opened 8 days ago by

New activity in OrionLLM/GRM-2.6-Plus-0628 8 days ago

Great model! Chat template issue [fixed].

#3 opened 17 days ago by

liked 2 models 9 days ago

sanjxz/Qwen-Fixed-Chat-Templates

Updated 22 days ago • 6

Moore2877/Qwen-Fixed-Chat-Templates-llamacpp

Updated 20 days ago • 21

liked a model 10 days ago

DavidAU/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP

Image-Text-to-Text • 28B • Updated about 8 hours ago • 26.1k • • 85

liked a model 12 days ago

nightmedia/Qwen3.6-27B-Architect-Polaris2-Fable-B-F451

Image-Text-to-Text • 28B • Updated 16 days ago • 3.87k • 23