GGUF models by mmnga-o

38 GGUF models quantized by mmnga-o on Hugging Face. Sizes run from 24M to 1420B parameters. The longest trained context is 1024K tokens.

Most downloadedNewestLargestLongest contextSmallest
QW
Qwen3 Swallow 8B RL v0.2
Qwen · mmnga-o
Params8.19B
Context40K
Q4_K_M5.0 GB
GGUF
LL
Llm Jp 4 8b Instruct
Llm · mmnga-o
Params8.59B
Context64K
Q4_K_M5.3 GB
GGUF
CA
CAT Thinking 8B
Cat · mmnga-o
Params8.19B
Context40K
Q4_K_M5.0 GB
Reasoning
PL
Plamo 3 Nict 2b Base
Plamo · mmnga-o
Params2.6B
Context4K
Q4_K_M1.7 GB
GGUF
NE
NVIDIA Nemotron Nano 9B v2 Japanese
Nemotron · mmnga-o
Params8.89B
Context1024K
Q4_K_M6.5 GB
GGUF
EL
ELYZA Thinking 1.0 llm jp 4 33b
Elyza · mmnga-o
Params33.2B
Context64K
Q4_K_M20.2 GB
Reasoning
EL
ELYZA Thinking 1.0 llm jp 4 32b a3b
Elyza · mmnga-o
Params32.1B
Context64K
Q4_K_M21.4 GB
Reasoning
CA
CAT Translate 1.4b
Cat · mmnga-o
Params1.41B
Context8K
Q4_K_M0.9 GB
GGUF
QW
Qwen3 Swallow 30B A3B RL v0.2
Qwen · mmnga-o
Params30.5B
Context40K
Q4_K_M18.6 GB
GGUF
CA
CAT Translate 0.8b
Cat · mmnga-o
Params793M
Context8K
Q4_K_M0.5 GB
GGUF
LL
Llm Jp 4 8b Thinking
Llm · mmnga-o
Params8.59B
Context64K
Q4_K_M5.3 GB
ReasoningTool use
LL
ArrowCanaria Llama 8B SFT v0.1
Llama · mmnga-o
Params8.03B
Context128K
Q4_K_M4.9 GB
GGUF
GP
GPT OSS Swallow 20B RL v0.1
gpt-oss · mmnga-o
Params20.9B
Context128K
Q4_K_M15.8 GB
Tool use
QW
Qwen3 Swallow 30B A3B SFT v0.2
Qwen · mmnga-o
Params30.5B
Context40K
Q4_K_M18.6 GB
GGUF
QW
Qwen3 Swallow 8B SFT v0.2
Qwen · mmnga-o
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
LL
Llm Jp 4 Kappa 32b A3B v0.1
Llm · mmnga-o
Params32.1B
Context64K
Q4_K_M21.4 GB
GGUF
PL
Plamo 3 Nict 8b Base
Plamo · mmnga-o
Params8.09B
Context4K
Q4_K_M5.0 GB
GGUF
CA
CAT Translate 7b
Cat · mmnga-o
Params7.49B
Context32K
Q4_K_M4.5 GB
GGUF
QW
Qwen3.5 0.8B JP
Qwen · mmnga-o
Params752M
Context256K
Q4_K_M0.5 GB
ReasoningTool useVision
LL
Llm Jp 4 32b A3B Thinking
Llm · mmnga-o
Params32.1B
Context64K
Q4_K_M21.4 GB
ReasoningTool use
LL
Llm Jp 4 33b Thinking
Llm · mmnga-o
Params33.2B
Context64K
Q4_K_M20.2 GB
ReasoningTool use
MI
miotts cpp
Miotts · mmnga-o
Params—
Context—
Smallest0.0 GB
GGUF
GP
GPT OSS Swallow 120B RL v0.1
gpt-oss · mmnga-o
Params117B
Context128K
Q4_K_M87.9 GB
Tool use
QW
Qwen 3 VL Ricoh 8B 20260227
Qwen · mmnga-o
Params8.19B
Context256K
Q4_K_M5.0 GB
ReasoningTool useVision
LF
LFM2.5 1.2B JP
Lfm · mmnga-o
Params1.17B
Context125K
Q4_K_M0.7 GB
Tool use
AR
ArrowIdeative 13b NeoBase ZERO Llm Jp v0.1
Arrowideative · mmnga-o
Params13.7B
Context4K
Q4_K_M8.3 GB
GGUF
EL
ELYZA Diffusion Instruct 1.0 Dream 7B
Elyza · mmnga-o
Params7.62B
Context128K
Q4_K_M4.7 GB
Tool use
KI
Kimi-K3
Kimi · mmnga-o
Params737B
Context1024K
Smallest194.0 GB
ReasoningTool use
QW
ABEJA Qwen3 14B Agentic 256k v0.1
Qwen · mmnga-o
Params14.8B
Context256K
Q4_K_M9.0 GB
ReasoningTool use
RA
RakutenAI 3.0
Rakutenai · mmnga-o
Params671B
Context160K
Q4_K_M404.5 GB
Tool use

About mmnga-o on Hugging Face

mmnga-o has published 38 GGUF models that ggufmodels.com tracks, across families such as Qwen, Llm and Cat. The most downloaded are Qwen3 Swallow 8B RL v0.2, Llm Jp 4 8b Instruct and CAT Thinking 8B.

The quants mmnga-o publishes most often are Q5_K_M (31), Q4_K_M (31), Q3_K_M (31) and Q8_0 (30). Their Q4_K_M files range from 0.5 to 404.5 GB. Parameter counts run from 24M to 1420B, and the longest trained context window is 1024K tokens.

Frequently asked questions

What GGUF models has mmnga-o published?

38 models are listed here, mainly from the Qwen, Llm and Cat families. The most downloaded is Qwen3 Swallow 8B RL v0.2.

Which quants does mmnga-o publish?

The most common are Q5_K_M (31 models), Q4_K_M (31 models), Q3_K_M (31 models) and Q8_0 (30 models).

Keep browsing