GGUF models by mmnga
261 GGUF models quantized by mmnga on Hugging Face. Sizes run from 268M to 671B parameters. The longest trained context is 10240K tokens.
OR
Orion-14B-LongChat
Orion · mmnga
Params14.5B
Context195K
Q4_K_M8.8 GB
JA
japanese-stablelm-3b-4e1t-base
Japanese · mmnga
Params2.8B
Context4K
Q4_K_M1.7 GB
BA
baku 10b chat v2
Baku · mmnga
Params10.2B
Context32K
Q4_K_M6.2 GB
ST
stockmark-13b
Stockmark · mmnga
Params13.2B
Context2K
Q4_K_M8.0 GB
LL
llm-jp-1.3b-v1.0
Llm · mmnga
Params1.42B
Context2K
Q4_K_M0.9 GB
OR
Orion-14B-Chat-RAG
Orion · mmnga
Params14.5B
Context4K
Q4_K_M8.8 GB
DE
DeepSeek V3 Bf16
DeepSeek · mmnga
Params671B
Context160K
Smallest133.6 GB
RA
RakutenAI 7B
Rakutenai · mmnga
Params7.37B
Context32K
Q4_K_M4.5 GB
TO
tokyotech llm Swallow 7b plus hf
Tokyotech · mmnga
Params6.83B
Context4K
Q4_K_M4.1 GB
JA
japanese-large-lm-3.6b-instruction-sft
Japanese · mmnga
Params3.71B
Context2K
Smallest2.1 GB
TA
Tanuki-ZeRo
Tanuki · mmnga
Params13.1B
Context2K
Q4_K_M8.4 GB
RI
rinna nekomata 7b instruction
Rinna · mmnga
Params7.72B
Context32K
Q4_K_M4.9 GB
JA
japanese-large-lm-3.6b
Japanese · mmnga
Params3.71B
Context2K
Smallest2.1 GB
RI
rinna youri 7b
Rinna · mmnga
Params6.74B
Context4K
Q4_K_M4.1 GB
LI
lightblue karasu 1.1B
Lightblue · mmnga
Params1.1B
Context2K
Q4_K_M0.7 GB
BA
baku 10b
Baku · mmnga
Params10.2B
Context4K
Q4_K_M6.2 GB
JA
japanese stablelm base ja vocab beta 7b
Japanese · mmnga
Params6.88B
Context4K
Q4_K_M4.2 GB
SU
suzume-poc
Suzume · mmnga
Params2.51B
Context8K
Q4_K_M1.6 GB
BA
baku 13b v2
Baku · mmnga
Params13.5B
Context4K
Q4_K_M8.2 GB
DE
DeepSeek V3 0324 Experts Pertok 4
DeepSeek · mmnga
Params671B
Context160K
Q4_K_M404.4 GB
PH
Phi 4 Reasoning
Phi · mmnga
Params14.7B
Context32K
Q4_K_M9.1 GB