GGUF models that fit in 4 GB

79024 GGUF language models have at least one quant whose estimated memory use is 4 GB or less at 8K context. Each card shows the largest quant that fits. Sizes run from 1M to 20260731B parameters. The longest trained context is 1280000K tokens.

Most downloadedNewestLargestLongest contextSmallest
QW
Qwen3 Vl 4b Heretic
Qwen · DreamFast
Params4B
Context256K
Best fitQ3_K_M 3.8 GB
Vision
GE
Gemma 3 4b It Qat
Gemma · bartowski
Params3.88B
Context128K
Best fitIQ4_NL 4.0 GB
Vision
LI
LightOnOCR 3 0.8B
Lightonocr · Prathat
Params752M
Context256K
Best fitF16 2.4 GB
Vision
DE
DeepSeek R1 Distill Qwen 7B Abliterated v2
DeepSeek · mradermacher
Params7.62B
Context128K
Best fitQ2_K 4.0 GB
Reasoning
QW
Qwen3.8 27B
Qwen · mradermacher
Params27.3B
Context256K
Q4_K_M16.8 GB
GGUF
LL
Llama 3.2 1B Instruct
Llama · DavidAU
Params1.24B
Context128K
Best fitQ6_K 1.8 GB
Roleplay
LI
LightOnOCR 3 0.8B
Lightonocr · prithivMLmods
Params752M
Context256K
Best fitBF16 2.4 GB
Vision
OC
Occamy 1.0
Occamy · bartowski
Params34.7B
Context256K
Q4_K_M22.3 GB
Tool useVision
QW
Bella Bartender Qwen3.5 2b
Qwen · mradermacher
Params1.88B
Context256K
Best fitQ6_K 2.5 GB
ReasoningTool use
SM
SmolLM 1.7B
SmolLM · bartowski
Params1.71B
Context2K
Best fitQ8_0 3.9 GB
Tool use
GL
GLM OCR
GLM · mradermacher
Params891M
Context128K
Best fitF16 2.9 GB
Vision
QW
Qwen2.5 VL 32B Instruct Heretic
Qwen · mradermacher
Params32.8B
Context125K
Q4_K_M19.9 GB
CodingVision
RE
Reggie v2 7b
Reggie · mradermacher
Params7.62B
Context128K
Best fitQ2_K 4.0 GB
ReasoningVision
QW
qwen3 tts 0 6b custom voice mix precision
Qwen · Arm
Params83M
Context8K
Best fitQ8_0 1.3 GB
GGUF
HU
HuatuoGPT O1 7B
Huatuogpt · bartowski
Params7.62B
Context32K
Best fitQ2_K 4.0 GB
Reasoning
ZD
TaichuAI/ZDTaichu5.0-9B
Zdtaichu · ngquocvinh
Params8.95B
Context256K
Best fitQ1_0 3.5 GB
ReasoningVision
QW
Qwen2.5 3B Instruct
Qwen · ThomasBaruzier
Params3.09B
Context32K
Best fitQ6_K 3.3 GB
GGUF
IN
InternVL3_5 30B A3B
Internvl · bartowski
Params30.5B
Context40K
Q4_K_M18.6 GB
Vision
QW
DarkIdol Qwen3.8 27B v1.1
Qwen · mradermacher
Params27.3B
Context256K
Q4_K_M16.8 GB
ReasoningRoleplayVision
RU
RusMind 4.7 Momentum
Rusmind · DICPlay
Params4.65B
Context128K
Q4_K_M3.4 GB
ReasoningTool use
0G
0GM 1.0 35B A3B 0427
mradermacher
Params34.7B
Context256K
Q4_K_M21.2 GB
GGUF
QW
PAW 27B X3.1
Qwen · lackonendes
Params4.86B
Context256K
Best fitQ2_K 1.4 GB
GGUF
DA
L3 Dark Planet 8B
Dark · DavidAU
Params8B
Context8K
Best fitIQ2_XXS 4.0 GB
Roleplay
HE
Hemmingway 1 Uncensored Heretic
Hemmingway · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
GGUF
OL
Olmo 3.1 32B Think
OLMo · bartowski
Params32.2B
Context64K
Q4_K_M19.5 GB
Reasoning
EX
EXAONE-4.0-1.2B
Exaone · LGAI-EXAONE
Params1.28B
Context64K
Best fitBF16 3.6 GB
GGUF
DE
DeepSeek R1 0528 Qwen3 8B heretic
DeepSeek · richardyoung
Params8.2B
Context128K
Q4_K_M5.0 GB
Reasoning
PA
MS3.2 PaintedFantasy Visage v2 33B
Paintedfantasy · bartowski
Params33.6B
Context128K
Q4_K_M20.4 GB
GGUF
QW
Qwen2.5 Coder 7B Instruct Heretic
Qwen · mradermacher
Params7.62B
Context32K
Best fitQ2_K 4.0 GB
Coding
DA
Davinci Tool 30B
Davinci · mradermacher
Params30.5B
Context256K
Q4_K_M18.6 GB
Tool use

Other memory sizes

Keep browsing