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
DE
DeepSeek R1 0528 Qwen3 8B
DeepSeek · mradermacher
Params8.19B
Context128K
Q4_K_M5.0 GB
ReasoningTool use
BE
34b beta i1
Beta · mradermacher
Params34.4B
Context195K
Q4_K_M20.7 GB
GGUF
ME
Megamind v2 VL High
Megamind · mradermacher
Params8.19B
Context256K
Q4_K_M5.0 GB
ReasoningTool useVision
ST
L3-8B-Stheno-v3.1
Stheno · duyntnet
Params8.03B
Context8K
Q4_K_M4.9 GB
GGUF
LO
LongSeeker 30B SFT
Longseeker · mradermacher
Params30.5B
Context256K
Q4_K_M18.6 GB
ReasoningTool use
QW
Qwen3 8B Base
Qwen · mradermacher
Params8.19B
Context32K
Q4_K_M5.0 GB
ReasoningTool use
ST
StepFun Formalizer 7B
Stepfun · mradermacher
Params7.62B
Context128K
Q4_K_M4.7 GB
Tool use
JI
Jina Embeddings v4 Vllm Retrieval
Jina · mradermacher
Params3.1B
Context125K
Q4_K_M1.9 GB
Embedding
GO
Goetia 24B v1.4
Goetia · mradermacher
Params23.6B
Context128K
Q4_K_M14.3 GB
RoleplayTool use
QW
Nsfwvision V4_Qwen3.5 9b Sft
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
Vision
TI
Tiamat 7b 1.1 Dpo
Tiamat · duyntnet
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
EC
EchoSwift 700M
Echoswift · mradermacher
Params742M
Context125K
Q4_K_M0.5 GB
Tool use
MI
Ministral 3 14B Reasoning 2512
Mistral · mradermacher
Params13.5B
Context256K
Q4_K_M8.2 GB
Reasoning
MI
Mistral Small 3.1 24B Instruct 2503
Mistral · mradermacher
Params23.6B
Context128K
Q4_K_M14.3 GB
GGUF
LL
Esper-70b
Llama · backyardai
Params69B
Context4K
Q4_K_M41.4 GB
Coding
GL
Austral GLM4 Winton
GLM · mradermacher
Params32.6B
Context32K
Q4_K_M19.7 GB
ReasoningRoleplay
OC
OctoThinker 8B Hybrid Base
Octothinker · mradermacher
Params8B
Context128K
Q4_K_M4.9 GB
GGUF
HO
K2-Horizon-0.9B
Horizon · mradermacher
Params1.08B
Context128K
Q4_K_M0.7 GB
Reasoning
AG
Agens Pilot
Agens · mradermacher
Params27B
Context1024K
Q4_K_M16.5 GB
CodingVision
HO
K2 Horizon 0.9B Heretic
Horizon · mradermacher
Params1.08B
Context128K
Q4_K_M0.7 GB
GGUF
QW
Qwen2.5 Microsoft NextCoder Olympic Instruct FUSED CODER Fast 11B
Qwen · mradermacher
Params10.9B
Context32K
Q4_K_M6.6 GB
CodingReasoningTool use
LL
Llama 3.1 8B Instruct Heretic
Llama · mradermacher
Params8B
Context128K
Q4_K_M4.9 GB
GGUF
GL
GLM 4.7 Flash Abliteratex
GLM · mradermacher
Params30B
Context198K
Q4_K_M18.1 GB
GGUF
GE
Gemma 3 12b It GLM Flash 4.7 Heretic Thinking Uncensored GRANDE
Gemma · mradermacher
Params12.8B
Context128K
Q4_K_M7.9 GB
Reasoning
BA
Baichuan M2 32B Abliterated
Baichuan · mradermacher
Params32.8B
Context128K
Q4_K_M19.9 GB
ReasoningTool use
FA
Falcon3 10B Instruct
Falcon · mradermacher
Params10.3B
Context32K
Q4_K_M6.3 GB
GGUF
FI
Finch 8B KTO
Finch · mradermacher
Params8.19B
Context128K
Q4_K_M5.0 GB
ReasoningTool use
MI
Mistral 7b Instruct v0.2 Heretic
Mistral · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
LI
A.X-4.0-Light
Light · Mungert
Params7.26B
Context16K
Q4_K_M4.5 GB
GGUF
QW
Qwen3.5 9B Uncensored Safetensors
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
GGUF

Other quants

Keep browsing