Michael Radermacher
mradermacher
AI & ML interests
Trying to fill the gaps with static and imatrix GGUF quantizations, especially of older, possibly obscure models, or big models that are hard to quantize/run. Feel free to suggest models (e.g. by opening an issue on the model_requests model)
Organizations
None yet
mradermacher's activity
Tested Fimburs11V3.i1-IQ3_XXS.gguf - Unsuable
#1 opened about 3 hours ago
by
chew1976
SanjiWatsuki/Silicon-Maid-7B high perplexity
16
#70 opened 6 days ago
by
yttria
Thanks and Question about quanting...
16
#1 opened about 2 months ago
by
DavidAU
nbeerbower/llama-3-SNAMD-8B
1
#79 opened about 9 hours ago
by
snoopynoob
archangel_sft-kto_llama30b ?
6
#78 opened 1 day ago
by
Elfrino
Usage info
1
#1 opened about 13 hours ago
by
catinthebag
requires newer version of koboldcpp
3
#1 opened about 14 hours ago
by
lemon07r
model.safetensors.index.json missing?
2
#1 opened 1 day ago
by
mradermacher
Model info
1
#1 opened 1 day ago
by
colourspec
Imatrix quants
1
#1 opened 1 day ago
by
hirosenn
GGUF Defect? | unknown pre-tokenizer type: 'smaug-bpe'
1
#1 opened 2 days ago
by
Skydea
Quill 72b (safetensors?)
2
#77 opened 2 days ago
by
leafspark
cognitivecomputations/dolphin-2.9.2-mixtral-8x22b
2
#76 opened 2 days ago
by
OrangeApples
hikikomoriHaven/llama3-8b-hikikomori-v0.3
1
#75 opened 3 days ago
by
woofwolfy
https://huggingface.co/ChuckMcSneed/dolphin-2.9.1-dbrx-llamacppfixed
1
#74 opened 3 days ago
by
nicoboss
More experiments 😸
1
#73 opened 4 days ago
by
saishf
Could you quantize those please?
1
#72 opened 4 days ago
by
xxx777xxxASD
JSL-MedPhi2
5
#1 opened 18 days ago
by
eleius
https://huggingface.co/fblgit/UNA-ThePitbull-21.4-v1
4
#71 opened 6 days ago
by
Dsol58
Model Request - https://huggingface.co/fblgit/UNA-ThePitbull-21.4-v1
5
#69 opened 6 days ago
by
USM-Valor
Japanese iMatrix dataset
1
#1 opened 5 days ago
by
keitokei1994
Anyway to make gguf's? :)
5
#1 opened 18 days ago
by
vdruts
sweet nothings
2
#1 opened 5 days ago
by
mradermacher
Imatrix quants?
1
#1 opened 6 days ago
by
hirosenn
Thank you!
3
#1 opened 7 days ago
by
HiroseKoichi
Request for imatrix version
1
#1 opened 7 days ago
by
xalkoi
thanks bro!
1
#1 opened 7 days ago
by
Guilherme34
https://huggingface.co/jondurbin/bagel-dpo-20b-v04-llama
3
#67 opened 7 days ago
by
Cran-May
ryzen88/Llama-3-70b-Arimas-story-RP-V1
1
#68 opened 7 days ago
by
ryzen88
Requesting smaller i Quants of Goliath 120B Longlora version OR assistance doing it on my own PC
20
#42 opened about 1 month ago
by
GoldenSun3DS
https://huggingface.co/PJMixers/LLaMa-3-CursedStock-v1.6-8B
1
#66 opened 8 days ago
by
xzuyn
im-request: Undi95/Llama-3-Chatty-2x8B (https://huggingface.co/Undi95/Llama-3-Chatty-2x8B)
3
#65 opened 8 days ago
by
xpgx1
Hastagaras/Halu-8B-Llama3-v0.3
2
#64 opened 8 days ago
by
Hastagaras
Could you quant maldv/badger-iota-llama-3-8b?
5
#63 opened 9 days ago
by
maldv
Stop token fix?
2
#1 opened 10 days ago
by
americancookie
Help me quantize model
16
#57 opened 12 days ago
by
Alsebay
🚩 Report
1
#1 opened 11 days ago
by
FabioA810
jukofyork/Dark-Miqu-120B
1
#62 opened 11 days ago
by
colourspec
Get well soon,man!!
1
#61 opened 12 days ago
by
OmnicromsBrain
Quantization request
2
#60 opened 12 days ago
by
xxx777xxxASD
missing tokenizer?
2
#1 opened 12 days ago
by
mradermacher
Deep Miqu 103b- https://huggingface.co/jukofyork/Deep-Miqu-103B
2
#56 opened 13 days ago
by
Dsol58
MAmmoTH2 8x7B and MAmmoTH2 8x7B Plus
1
#58 opened 12 days ago
by
Erilaz
request: larger context (64/256) window on uncensored models
1
#59 opened 12 days ago
by
exquisiteart