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
IN
Internlm2 Math Plus 7b
InternLM · mradermacher
Params7.74B
Context8K
Best fitIQ2_XS 4.0 GB
GGUF
QW
Qwen3.5 40B Claude 4.6 Opus Deckard Heretic Uncensored Thinking
Qwen · FrosT-LyRo
Params39.1B
Context256K
Q4_K_M23.9 GB
CodingReasoningRoleplay
QW
Qwen3-Vl-2B-Thinking-1M
Qwen · unsloth
Params1.72B
Context977K
Best fitQ8_K_XL 3.8 GB
ReasoningVision
LL
Collaiborator-MEDLLM-Llama-3-8B-v2
Llama · mradermacher
Params8B
Context8K
Best fitIQ2_XXS 4.0 GB
GGUF
TH
Thaki ASR KO Dialects v1 i1
Thaki · mradermacher
Params1.72B
Context64K
Best fitQ6_K 2.9 GB
GGUF
NO
nomic ai nomic embed code
Nomic · bartowski
Params7.1B
Context32K
Best fitIQ3_XXS 3.9 GB
CodingEmbedding
EK
Ekbasis Skopos 27B
Ekbasis · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
GGUF
AI
aiXcompact 4B
Aixcompact · mradermacher
Params4.21B
Context256K
Best fitQ3_K_L 4.0 GB
CodingReasoningTool use
IN
Internlm2 Math Plus 1_8b
InternLM · mradermacher
Params1.89B
Context8K
Best fitQ6_K 2.9 GB
GGUF
SM
SmolLM2 360M Instruct
SmolLM · unsloth
Params362M
Context8K
Best fitF16 1.6 GB
GGUF
QW
Huihui Qwen3 30B A3B Thinking 2507 Abliterated
Qwen · mradermacher
Params30.5B
Context256K
Q4_K_M18.6 GB
Reasoning
MI
Magistral Small 2509
Mistral · bartowski
Params23.6B
Context128K
Q4_K_M14.3 GB
Reasoning
SH
Shaylah Darija 4b Merged
Shaylah · mradermacher
Params4.21B
Context256K
Best fitQ3_K_L 4.0 GB
ReasoningTool use
ZO
Zoro i1
Zoro · mradermacher
Params4.84B
Context256K
Best fitIQ3_XS 4.0 GB
ReasoningTool use
NE
Nex N2.5 Pro
Nex · bartowski
Params396B
Context256K
Q4_K_M250.8 GB
Vision
TR
TranslatePsy AfriSLM 0.8B
Translatepsy · mradermacher
Params1.01B
Context256K
Best fitQ6_K 1.8 GB
ReasoningTool use
GE
Gemma 3 4B it Uncensored
Gemma · lemuralabs
Params3.88B
Context128K
Best fitQ3_K_M 3.7 GB
GGUF
DO
DoItYourself v1 2B
Doityourself · theprint
Params1.94B
Context256K
Best fitQ8_0 3.0 GB
GGUF
QW
Twi Qwen v1 Merged
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningTool use
ME
MechVL 4B RL
Mechvl · mradermacher
Params4.41B
Context256K
Best fitQ3_K_M 4.0 GB
ReasoningVision
SO
SOR ColdBrew 12B Think RP Base
Sor · mradermacher
Params12.2B
Context1000K
Q4_K_M7.5 GB
ReasoningRoleplay
MI
Hermes-2-Pro-Mistral-7B
Mistral · mradermacher
Params7.24B
Context32K
Best fitIQ2_S 3.9 GB
Tool use
QW
Qwen3.8 27B OBLITERATED Mythos Class Agentic
Qwen · mradermacher
Params27.3B
Context256K
Q4_K_M16.8 GB
ReasoningTool use
GP
Gpt Oss 20b Unsloth Puzzle V2
gpt-oss · mradermacher
Params20.9B
Context128K
Q4_K_M15.8 GB
Reasoning
TR
TranslatePsy AfriSLM 4B
Translatepsy · mradermacher
Params4.84B
Context256K
Best fitQ3_K_S 4.0 GB
ReasoningTool use
ON
OneJev 0.8B
Onejev · bartowski
Params1.01B
Context256K
Best fitBF16 2.9 GB
Vision
HO
K2 Horizon MoVA 36B A4B
Horizon · ngquocvinh
Params37.4B
Context512K
Q4_K_M22.4 GB
GGUF
LL
Llama Leonida
Llama · mradermacher
Params8.2B
Context128K
Best fitIQ1_M 3.8 GB
Tool use
ST
StartLux-Decision-v1.1-0.8B-Instant
Startlux · startlux-models
Params752M
Context256K
Best fitQ8_0 1.7 GB
Vision
AN
Andromeda 8B Coder i1
Andromeda · mradermacher
Params8B
Context128K
Best fitIQ2_XXS 4.0 GB
CodingTool use

Other memory sizes

Keep browsing