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 8B Heretic
Qwen · mradermacher
Params8.19B
Context40K
Best fitIQ1_M 4.0 GB
GGUF
GE
Gemma 3 it 4B Uncensored DBL X
Gemma · DavidAU
Params4.26B
Context128K
Best fitQ3_K_S 3.9 GB
Roleplay
GP
GPT OSS Cybersecurity 20B Merged Heretic
gpt-oss · mradermacher
Params20.9B
Context128K
Q4_K_M15.8 GB
GGUF
YI
Hermes-2.5-Yi-1.5-9B-Chat
Yi · mradermacher
Params8.83B
Context4K
Best fitIQ2_XS 4.0 GB
Tool use
QW
dolphin-2.9.2-qwen2-7b
Qwen · dphn
Params7.62B
Context128K
Best fitQ2_K 4.0 GB
GGUF
GE
Gemma 3 270m It
Gemma · bartowski
Params268M
Context32K
Best fitBF16 1.2 GB
GGUF
SP
SpydazWebAI_Human_AGI
Spydazwebai · mradermacher
Params7.24B
Context512K
Best fitIQ2_S 3.9 GB
Coding
GE
gemma 4 e2b yemeni arabic assistant
Gemma · yemeni-ai-lab
Params4.63B
Context128K
Best fitTQ2_0 3.9 GB
Vision
MI
Infinity Instruct 7M 0729 Mistral 7B
Mistral · mradermacher
Params7.24B
Context32K
Best fitIQ2_S 3.9 GB
GGUF
GR
granite-3b-code-base
Granite · mradermacher
Params3.48B
Context2K
Best fitIQ1_S 4.0 GB
Coding
CA
CantoneseLLM v2.0 8B Thinking Chat Vector Merged
Cantonesellm · mradermacher
Params8.19B
Context40K
Best fitIQ1_M 4.0 GB
Reasoning
IN
Index Nailong 9B
Index · mradermacher
Params8.95B
Context224K
Q4_K_M5.6 GB
GGUF
MA
mamba 130m hf
Mamba · QuantFactory
Params129M
Context1024K
Best fitQ8_0 0.7 GB
GGUF
HY
HybridAgent 14B Agentic
Hybridagent · bartowski
Params14.8B
Context32K
Q4_K_M9.3 GB
ReasoningTool use
VI
T VisStar 7B v0.1 i1
Visstar · mradermacher
Params7.29B
Context32K
Best fitIQ2_S 3.9 GB
GGUF
VI
Vietnamese 4B Chat abliterated
Vietnamese · N14QM
Params4.2B
Context256K
Best fitQ3_K_L 4.0 GB
Vision
GE
Gemma 2B Uncensored
Gemma · lemuralabs
Params2.51B
Context8K
Best fitQ8_0 3.3 GB
Coding
QW
Huihui Qwen3.5 9B Claude 4.6 Opus Abliterated Heretic
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
Reasoning
QW
Qwen2.5 7B Instruct abliterated v2
Qwen · kepom
Params7.6B
Context32K
Best fitQ2_K 4.0 GB
GGUF
QW
Qwen3.5 2B EmperoAI Qwen3.8 Distill Heretic Abliterated
Qwen · mradermacher
Params1.88B
Context256K
Best fitQ8_0 2.9 GB
ReasoningTool use
LL
Llama3-ChatQA-1.5-8B
Llama · legraphista
Params8B
Context8K
Best fitIQ2_XXS 4.0 GB
GGUF
TM
Tmax v1.1 9b
Tmax · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
GGUF
ED
vllm-d3-edge
Edge · ggml-org
Params748M
Context256K
Best fitBF16 2.2 GB
Vision
AL
ALoDLM 8B
Alodlm · mradermacher
Params8.19B
Context40K
Best fitIQ1_M 4.0 GB
Coding
QW
Logics SWE Qwen3.6 27B
Qwen · mradermacher
Params26.9B
Context256K
Q4_K_M16.5 GB
CodingReasoningTool use
FR
FrogNano 4B 2609
Frognano · webmp3
Params4.33B
Context256K
Smallest2.0 GB
Vision
QW
Qwen3 1.7B
Qwen · gaianet
Params2.03B
Context40K
Best fitQ8_0 3.6 GB
GGUF
JA
Jan-v3-4b-base-instruct
Jan · janhq
Params4.4B
Context256K
Best fitQ3_K_M 4.0 GB
Coding
WE
WeVisOne VL 2B
Wevisone · mradermacher
Params2.03B
Context256K
Best fitQ8_0 3.6 GB
Tool useVision
GR
Granite 4.2 3b
Granite · bartowski
Params3.66B
Context128K
Best fitQ5_K_L 4.0 GB
ReasoningTool use

Other memory sizes

Keep browsing