blank reg
blankreg
AI & ML interests
None yet
Recent Activity
new activity 6 minutes ago
wangzhang/Qwen3.5-27B-abliterated:Thanks! liked a model 20 minutes ago
p-e-w/gpt-oss-20b-heretic-ara-v4 liked a model 22 minutes ago
wangzhang/Qwen3.5-27B-abliteratedOrganizations
None yet
llama.cpp
โ 1
#1 opened about 23 hours ago
by
blankreg
PIQA acc_norm
๐ 1
3
#1 opened 7 days ago
by
blankreg
mistral-vibe
1
#2 opened 2 days ago
by
blankreg
best qwen3.5 4b heretic so far
๐ 1
2
#1 opened about 1 month ago
by
TerrorJack
DeepSeek-OCR-2 ?
1
#1 opened 13 days ago
by
blankreg
Testing IQ4_NL
13
#12 opened about 2 months ago
by
shewin
fused up + gate conversions
๐ 1
1
#5 opened 29 days ago
by
blankreg
New update?
#17 opened about 1 month ago
by
blankreg
trohrbaugh/GLM-4.7-Flash-heretic
1
#1953 opened about 1 month ago
by
blankreg
trohrbaugh/Seed-OSS-36B-Instruct-heretic
2
#1954 opened about 1 month ago
by
blankreg
mmproj F32 only?
โ 1
#27 opened about 1 month ago
by
blankreg
trohrbaugh/Qwen3-Coder-Next-heretic
1
#1936 opened about 1 month ago
by
blankreg
Suggestion
#1 opened about 1 month ago
by
blankreg
Q3_K_M (112 GB) is bigger than Q3_K_XL (104 GB)?
4
#8 opened 6 months ago
by
rtzurtz
Update: Should now be Fixed - Bug in UD-Q4_K_XL recipe using MXFP4 for attn tensors and experts?
๐ 8
26
#5 opened about 1 month ago
by
ubergarm
Differences?
5
#2 opened about 1 month ago
by
blankreg
GGUF
๐ 1
#17 opened about 1 month ago
by
blankreg
Fix for Nemotron forced thinking (can be toggled on and off)
#9 opened 3 months ago
by
blankreg