GGUF models by SandLogicTechnologies

98 GGUF models quantized by SandLogicTechnologies on Hugging Face. Sizes run from 164M to 70.6B parameters. The longest trained context is 1024K tokens.

Most downloadedNewestLargestLongest contextSmallest
TR
Translategemma 4b It
Translategemma · SandLogicTechnologies
Params3.88B
Context128K
Q4_K_M2.5 GB
GGUF
FR
Fara1.5 4B
Frognano · SandLogicTechnologies
Params4.2B
Context256K
Smallest2.2 GB
CodingReasoning
FR
FrogNano 4B 2609
Frognano · SandLogicTechnologies
Params4.33B
Context256K
Smallest2.2 GB
CodingReasoningTool use
QW
Qwen3 0.6B
Qwen · SandLogicTechnologies
Params752M
Context32K
Q4_K_M0.5 GB
ReasoningTool use
DE
DeepSeek OCR 2
DeepSeek · SandLogicTechnologies
Params2.93B
Context8K
Smallest1.4 GB
GGUF
CH
Chandra Ocr 2
Chandra · SandLogicTechnologies
Params4.96B
Context256K
Smallest2.5 GB
ReasoningTool useVision
PA
PaddleOCR vl 1.6
Paddleocr · SandLogicTechnologies
Params467M
Context128K
Smallest0.3 GB
Vision
MA
mage vl
Mage · SandLogicTechnologies
Params4.41B
Context256K
Smallest2.1 GB
Vision
MI
MiniCPM OCR Jev
MiniCPM · SandLogicTechnologies
Params1.88B
Context256K
Smallest1.1 GB
ReasoningTool useVision
OR
Ornith-1.5-35B
Ornith · SandLogicTechnologies
Params35.5B
Context256K
Smallest15.8 GB
CodingReasoningTool use
QW
Qwen3.8 27B
Qwen · SandLogicTechnologies
Params27.3B
Context256K
Smallest12.8 GB
ReasoningTool useVision
MI
MiniCPM V 4.7 35B A3B
MiniCPM · SandLogicTechnologies
Params34.7B
Context256K
Smallest15.4 GB
ReasoningTool useVision
WE
WeVisOne VL 4B
Wevisone · SandLogicTechnologies
Params4.42B
Context256K
Smallest2.1 GB
Vision
QW
Qwen2.5 VL 7B Instruct
Qwen · SandLogicTechnologies
Params14.8B
Context40K
Q4_K_M9.0 GB
ReasoningTool useVision
GP
gpt oss 20b
gpt-oss · SandLogicTechnologies
Params20.9B
Context128K
Smallest12.2 GB
ReasoningTool use
GR
Granite Embedding 311m Multilingual R2
Granite · SandLogicTechnologies
Params312M
Context8K
Smallest0.2 GB
CodingEmbedding
ME
Medgemma 4b It
Medgemma · SandLogicTechnologies
Params3.88B
Context128K
Q4_K_M2.5 GB
ReasoningVision
MI
MiniCPM5 2B
MiniCPM · SandLogicTechnologies
Params2.52B
Context128K
Smallest1.2 GB
ReasoningTool use
LF
LFM2.5 1.2B Instruct
Lfm · SandLogicTechnologies
Params1.17B
Context125K
Smallest0.6 GB
ReasoningTool use
GR
Granite Docling 258M
Granite · SandLogicTechnologies
Params164M
Context8K
Smallest0.1 GB
Vision
MI
MiniCPM V 4.6
MiniCPM · SandLogicTechnologies
Params752M
Context256K
Smallest0.5 GB
ReasoningTool useVision
DE
DeepSeek Coder V2 Lite Instruct
DeepSeek · SandLogicTechnologies
Params15.7B
Context160K
Q4_K_M10.4 GB
Coding
GL
GLM 4.7 Flash
GLM · SandLogicTechnologies
Params29.9B
Context198K
Smallest13.2 GB
CodingReasoningTool use
DO
dots.mocr
Dots · SandLogicTechnologies
Params1.78B
Context128K
Smallest0.9 GB
Vision
LO
LocateAnything 3B
Locateanything · SandLogicTechnologies
Params3.4B
Context32K
Smallest1.6 GB
Tool useVision
OR
Ornith 1.0 9B
Ornith · SandLogicTechnologies
Params8.95B
Context256K
Smallest4.4 GB
CodingReasoningTool use
LL
Llama 3.2 3B Instruct
Llama · SandLogicTechnologies
Params3.21B
Context128K
Q4_K_M2.0 GB
Tool use
LA
laya
Laya · SandLogicTechnologies
Params421M
Context8K
Smallest0.2 GB
GGUF
GE
Gemma 4 E4B It
Gemma · SandLogicTechnologies
Params7.52B
Context128K
Smallest4.7 GB
ReasoningTool useVision
GE
gemma-2-9b-it
Gemma · SandLogicTechnologies
Params9.24B
Context8K
Q4_K_M5.8 GB
GGUF

About SandLogicTechnologies on Hugging Face

SandLogicTechnologies has published 98 GGUF models that ggufmodels.com tracks, across families such as Llama, Qwen and Gemma. The most downloaded are Translategemma 4b It, Fara1.5 4B and FrogNano 4B 2609.

The quants SandLogicTechnologies publishes most often are Q4_K_M (56), Q5_K_M (52), IQ4_XS (45) and IQ4_NL (40). Their Q4_K_M files range from 0.2 to 42.5 GB. Parameter counts run from 164M to 70.6B, and the longest trained context window is 1024K tokens.

Frequently asked questions

What GGUF models has SandLogicTechnologies published?

98 models are listed here, mainly from the Llama, Qwen and Gemma families. The most downloaded is Translategemma 4b It.

Which quants does SandLogicTechnologies publish?

The most common are Q4_K_M (56 models), Q5_K_M (52 models), IQ4_XS (45 models) and IQ4_NL (40 models).

Keep browsing