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
QW
Qwen3 Gutenberg Encore 14B
Qwen · mradermacher
Params14.8B
Context32K
Q4_K_M9.0 GB
ReasoningTool use
AI
Aitana 7B S Base
Aitana · mradermacher
Params7.77B
Context8K
Q4_K_M4.9 GB
GGUF
QW
QwQ 35B Eureka Cubed i1
Qwen · mradermacher
Params34.7B
Context128K
Q4_K_M21.0 GB
ReasoningTool use
QW
QwQ 32B Abliterated
Qwen · mradermacher
Params32.8B
Context128K
Q4_K_M19.9 GB
ReasoningTool use
LL
suzume llama 3 8B japanese i1
Llama · mradermacher
Params8.03B
Context8K
Q4_K_M4.9 GB
GGUF
QW
Qwen3 VL 4B Instruct
Qwen · mradermacher
Params4.02B
Context256K
Q4_K_M2.5 GB
Tool useVision
BO
Boris 1.3 125M Instruct
Boris · opencerebral
Params124M
Context1K
Q4_K_M0.1 GB
GGUF
GE
Gemma 3 27b It Abliterated Normpreserve
Gemma · mradermacher
Params27B
Context128K
Q4_K_M16.5 GB
GGUF
PH
Phoenix 7B i1
Phoenix · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
TH
The_Elder v2
The · mradermacher
Params1.1B
Context2K
Q4_K_M0.7 GB
GGUF
ST
StarMix 7B slerp i1
Starmix · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
QW
OpenAgent
Qwen · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
LL
Llama 3.3 70B Instruct
Llama · ThomasBaruzier
Params70.6B
Context128K
Q4_K_M42.5 GB
GGUF
PH
Phi-4-mini-reasoning
Phi · Mungert
Params3.84B
Context128K
Q4_K_M2.5 GB
CodingReasoning
TR
Trinity-Nano-Preview
Trinity · Mungert
Params6.12B
Context128K
Q4_K_M4.0 GB
Tool use
PA
MS3.2 PaintedFantasy v4.1 24B Ultra Uncensored Heretic v2
Paintedfantasy · mradermacher
Params23.6B
Context128K
Q4_K_M14.3 GB
CodingRoleplay
FA
WizardLM Uncensored Falcon 7b
Falcon · mradermacher
Params7.22B
Context2K
Q4_K_M5.0 GB
GGUF
LL
LLaMA 2 7B 32K
Llama · mradermacher
Params6.74B
Context32K
Q4_K_M4.1 GB
GGUF
VE
Vesper Zenith 12B
Vesper · mradermacher
Params12.2B
Context128K
Q4_K_M7.5 GB
Roleplay
MI
Mistral Nemo Instruct 2407 Heretic Noslop MPOA
Mistral · mradermacher
Params12.2B
Context128K
Q4_K_M7.5 GB
GGUF
QW
Qwen3-VL-30B-A3B-Instruct
Qwen · Mungert
Params30.5B
Context256K
Q4_K_M19.0 GB
Tool useVision
QW
Qwen3 Yoyo V3 42B A3B Thinking TOTAL RECALL ST TNG II
Qwen · mradermacher
Params42.4B
Context986K
Q4_K_M25.7 GB
CodingReasoningTool use
MI
MiniMax M2.1 REAP 50
Minimax · mradermacher
Params116B
Context192K
Q4_K_M70.4 GB
CodingTool use
TU
DR Tulu No RLER 8B
Tulu · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
MI
Ae Calem Mistral 7b v0.1
Mistral · mradermacher
Params7.29B
Context32K
Q4_K_M4.4 GB
GGUF
EU
EuroLLM 9B Instruct Ungated
Eurollm · mradermacher
Params9.15B
Context4K
Q4_K_M5.6 GB
GGUF
GE
Gemma 4 12B It
Gemma · mradermacher
Params11.9B
Context256K
Q4_K_M7.4 GB
ReasoningTool use
GI
Giftige Blume 31B v1 StyleSwap
Giftige · mradermacher
Params32.1B
Context256K
Q4_K_M19.5 GB
ReasoningRoleplayTool use
MI
RenCoder Ministral 3 8B Instruct 2512
Mistral · mradermacher
Params8.49B
Context256K
Q4_K_M5.2 GB
CodingTool use
AP
Apertus 8B MeditronFO
Apertus · mradermacher
Params8.05B
Context64K
Q4_K_M5.1 GB
ReasoningTool use

Other quants

Keep browsing