IQ3_S GGUF models

A GGUF quantization format. Lower numbers mean smaller files and more quality loss. 45340 models have a IQ3_S file. Sizes run from 1M to 1030B parameters. The longest trained context is 102400K tokens.

Most downloadedNewestLargestLongest contextSmallest
QW
Qwen2.5 Coder 14B E2E Tests
Qwen · mradermacher
Params14.8B
Context32K
Q4_K_M9.0 GB
Coding
PY
Pythia 1.4b Deduped
Pythia · mradermacher
Params1.41B
Context2K
Q4_K_M0.9 GB
GGUF
QW
OpenChat 3.5 7B Qwen v2.0
Qwen · mradermacher
Params7.24B
Context8K
Q4_K_M4.4 GB
GGUF
MI
Ministral 3 8B Reasoning 2512 Esper3.1
Mistral · mradermacher
Params8.49B
Context256K
Q4_K_M5.2 GB
CodingReasoning
QW
Qwopus3.8 27B Flash V2 Apostate Uncensored
Qwopus · mradermacher
Params27.3B
Context256K
Q4_K_M16.8 GB
Vision
QW
Qwen2.5 7B Instruct
Qwen · RichardErkhov
Params7.62B
Context32K
Q4_K_M4.7 GB
Tool use
MI
Chinese Mixtral 8x7B
Mistral · RichardErkhov
Params46.9B
Context32K
Q4_K_M28.6 GB
GGUF
CA
cat-v1.0-13b
Cat · mradermacher
Params13B
Context4K
Q4_K_M7.9 GB
GGUF
VE
VeriLoop E2 GSQ RCO
Veriloop · tahaalam2009
Params27B
Context256K
Smallest8.5 GB
CodingVision
MI
Hermes-2-Pro-Mistral-7B-infinity
Mistral · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
Tool use
JA
Javi0.3 2B
Javi · mradermacher
Params1.94B
Context256K
Q4_K_M1.3 GB
ReasoningTool use
LL
gpt 4o distil Llama 3.3 70B Instruct PaperWitch heresy i1
Llama · mradermacher
Params70.6B
Context128K
Q4_K_M42.5 GB
GGUF
MI
OpenHermes 2.5 Mistral 7B
Mistral · RichardErkhov
Params7.24B
Context32K
Q4_K_M4.4 GB
Tool use
IN
OpenChat 3.5 7B InternLM v2.0
InternLM · mradermacher
Params7.24B
Context8K
Q4_K_M4.4 GB
GGUF
QW
FlashModel Qwen 3.6 35B A3B Maxed
Qwen · mradermacher
Params35.5B
Context256K
Q4_K_M21.7 GB
ReasoningTool use
MI
Mistral Syndicate 7B
Mistral · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
GGUF
NE
Nehuronion Dvt 9b
Nehuronion · mradermacher
Params9.24B
Context8K
Q4_K_M5.8 GB
GGUF
OR
Ornith 1.5 122B
Ornith · mradermacher
Params125B
Context256K
Q4_K_M75.8 GB
ReasoningTool use
GR
Granite 3b Code Instruct 128k
Granite · RichardErkhov
Params3.48B
Context125K
Q4_K_M2.1 GB
Coding
QW
Qwen3.6 35B A3B Uncensored Genesis Final Safetensors
Qwen · mradermacher
Params35.5B
Context256K
Q4_K_M21.7 GB
ReasoningTool useVision
BR
Breeze-13B-32k-Instruct-v1_0
Breeze · mradermacher
Params12.7B
Context32K
Q4_K_M7.7 GB
GGUF
QW
Qwen3 14B Abliterated
Qwen · mradermacher
Params14.8B
Context40K
Q4_K_M9.0 GB
GGUF
LA
Lambda 17b PcDPO
Lambda · mradermacher
Params16.8B
Context32K
Q4_K_M10.3 GB
Roleplay
OP
openPangu Embedded 7B V1.1
Openpangu · mradermacher
Params8.03B
Context32K
Q4_K_M4.9 GB
Embedding
TH
TherapyBeagle 11B v1 i1
Therapybeagle · mradermacher
Params10.7B
Context32K
Q4_K_M6.5 GB
GGUF
FA
Falcon H1 7B Instruct
Falcon · tiiuae
Params7.59B
Context256K
Q4_K_M4.6 GB
GGUF
EC
EchoVLM_v2
Echovlm · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningTool useVision
MA
MAmmoTH2-8x7B
Mammoth · mradermacher
Params46.7B
Context32K
Q4_K_M28.4 GB
GGUF
EU
Euryale 1.3 longLORA 70b rope8 32k fp16 i1
Euryale · mradermacher
Params69B
Context4K
Q4_K_M41.4 GB
GGUF
IN
Internlm2 Math Plus 7b
InternLM · mradermacher
Params7.74B
Context8K
Q4_K_M4.7 GB
GGUF

Other quants

Keep browsing