Q2_K_S GGUF models

A GGUF quantization format. Lower numbers mean smaller files and more quality loss. 19715 models have a Q2_K_S file. Sizes run from 2M to 671B parameters. The longest trained context is 10240K tokens.

Most downloadedNewestLargestLongest contextSmallest
GR
Granite Speech 5.0 2b
Granite · mradermacher
Params1.84B
Context16K
Q4_K_M1.1 GB
GGUF
WE
WeirdCompound v1.7 24b
Weirdcompound · mradermacher
Params23.6B
Context128K
Q4_K_M14.3 GB
GGUF
AS
ASearcher Local 14B
Asearcher · mradermacher
Params14.8B
Context128K
Q4_K_M9.0 GB
Tool use
GP
OpenAi GPT Oss 36B BrainStorm20x Uncensored
gpt-oss · mradermacher
Params36.6B
Context128K
Q4_K_M27.5 GB
CodingReasoning
LL
Llama3.1 Heretic Unsensored
Llama · mradermacher
Params8B
Context128K
Q4_K_M4.9 GB
CodingTool use
AS
Assistant_Pepe_32B
Assistant · mradermacher
Params32.8B
Context40K
Q4_K_M19.8 GB
GGUF
CY
Cyberdonia 24B v11
Cyberdonia · mradermacher
Params23.6B
Context128K
Q4_K_M14.3 GB
GGUF
OR
Orion 26B A4B v1
Orion · mradermacher
Params25.2B
Context256K
Q4_K_M16.8 GB
GGUF
GL
GLM 4.7 Flash REAP 39
GLM · mradermacher
Params19.1B
Context198K
Q4_K_M11.6 GB
CodingTool use
HO
Hornybot Mara POV
Hornybot · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
Roleplay
GL
CreativeSocialAgent_GLM_4.7_Flash
GLM · mradermacher
Params30B
Context198K
Q4_K_M18.1 GB
GGUF
MI
MistralPrism 24B
Mistral · Aratako
Params23.6B
Context128K
Q4_K_M14.3 GB
GGUF
GR
Huihui Granite 4.0 H Tiny Abliterated
Granite · mradermacher
Params6.94B
Context1024K
Q4_K_M4.2 GB
GGUF
GL
GLM 4.7 Flash Springdragon i1
GLM · mradermacher
Params30B
Context198K
Q4_K_M18.1 GB
GGUF
PY
PyThagorean 10B i1
Pythagorean · mradermacher
Params10.3B
Context32K
Q4_K_M6.3 GB
Coding
CY
Cydonia 24B v4.3 Heretic
Cydonia · mradermacher
Params23.6B
Context128K
Q4_K_M14.3 GB
Coding
LL
Dobby Unhinged Llama 3.3 70B
Llama · mradermacher
Params70.6B
Context128K
Q4_K_M42.5 GB
RoleplayTool use
MI
Ministral 3B Heresy Unfiltered
Mistral · mradermacher
Params3.43B
Context256K
Q4_K_M2.1 GB
GGUF
QW
Qwen3 4B
Qwen · mradermacher
Params4B
Context40K
Q4_K_M2.5 GB
GGUF
CA
Cajal 9b v2 Full
Cajal · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
GGUF
QW
CriticLeanGPT Qwen3 32B RL
Qwen · mradermacher
Params32.8B
Context40K
Q4_K_M19.8 GB
GGUF
QW
Nu2 Lupi Qwen 14B
Qwen · mradermacher
Params14.8B
Context32K
Q4_K_M9.0 GB
GGUF
GP
Gpt Oss 20b Claude 4 Distill
gpt-oss · mradermacher
Params20.9B
Context128K
Q4_K_M15.8 GB
GGUF
UR
Urdu Ghazal Generator
Urdu · mradermacher
Params7.62B
Context32K
Q4_K_M4.7 GB
Tool use
QW
MANGO Qwen3 Omni 30B A3B Instruct
Qwen · mradermacher
Params30.5B
Context64K
Q4_K_M18.6 GB
ReasoningTool useVision
QW
Qwen2.5 3B Instruct
Qwen · legraphista
Params3.09B
Context32K
Smallest0.8 GB
Tool use
QW
ThinkingCap Qwen3.8 27B Uncensored Heretic
Qwen · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
Reasoning
QW
Qwen3 Coder REAP 25B A3B i1
Qwen · mradermacher
Params25B
Context256K
Q4_K_M15.1 GB
Coding
GP
GPT OSS Swallow 20B RL v0.1
gpt-oss · mradermacher
Params20.9B
Context128K
Q4_K_M15.8 GB
Coding
QW
Qwen2.5 Omni 7B Abliterated
Qwen · mradermacher
Params7.62B
Context32K
Q4_K_M4.7 GB
GGUF

Other quants

Keep browsing