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.
QW
imajev 4b
Qwen · mindchain
Params4.2B
Context256K
Best fitQ3_K_L 4.0 GB
HU
Huihui OmniCoder 9B Abliterated
Huihui · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
QW
SummOrchestra Qwen3 8B GRPO BRLP SAMSUM
Qwen · mradermacher
Params8.19B
Context40K
Best fitIQ1_M 4.0 GB
QW
Qwen2.5 7B
Qwen · mradermacher
Params7.62B
Context128K
Best fitQ2_K 4.0 GB
MI
MiniCPM5 2B DSpark
MiniCPM · magnitudedev
Params324M
Context128K
Smallest0.7 GB
LF
LFM3-VL-2B
Lfm · AtomicChat
Params1.73B
Context125K
Best fitQ8_0 4.0 GB
VI
VibeThinker 3B
Vibethinker · mradermacher
Params3.09B
Context128K
Best fitQ6_K 3.3 GB
QW
Qwen3.5 9B Claude Fable 5xMerge
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
GL
GLM-ASR-Nano-2512
GLM · Mungert
Params1.59B
Context8K
Best fitBF16 3.4 GB
TO
TooManyMixRolePlay-7B-Story_V3.5
Toomanymixroleplay · mradermacher
Params7.24B
Context32K
Best fitIQ2_S 3.9 GB
LF
LFM2 1.2B Extract
Lfm · LiquidAI
Params1.17B
Context125K
Best fitF16 3.9 GB
QW
Ministral 3 3B Qwen3.8 Max Reasoning Distilled
Qwen · mradermacher
Params3.43B
Context256K
Best fitQ5_K_S 3.8 GB
GE
Gemma 3 1b It Heretic Extreme Uncensored Abliterated
Gemma · mradermacher
Params1B
Context32K
Best fitQ6_K 1.7 GB
WE
WeVisOne VL 4B
Wevisone · mradermacher
Params4.42B
Context256K
Best fitQ3_K_M 4.0 GB
NO
Noema 1.5 4B Fast
Noema · mradermacher
Params4.21B
Context256K
Best fitQ3_K_L 4.0 GB
MI
TowerInstruct Mistral 7B v0.2
Mistral · mradermacher
Params7.24B
Context32K
Best fitIQ2_S 3.9 GB
TO
ToriiGate 0.5
Toriigate · DraconicDragon
Params4.84B
Context256K
Best fitQ3_K_S 3.9 GB
LF
LFM2 VL 450M
Lfm · bartowski
Params354M
Context125K
Best fitBF16 1.7 GB
GL
glm-4-9b-chat
GLM · legraphista
Params9.4B
Context128K
Best fitIQ1_S 3.9 GB
DA
danube3-4B
Danube · h2oai
Params3.96B
Context8K
Best fitQ4_K_M 3.6 GB
PY
Pythia 410m
Pythia · mradermacher
Params405M
Context2K
Best fitF16 2.1 GB
NO
Nock Coder 1.5b
Nock · mradermacher
Params1.54B
Context32K
Best fitQ6_K 2.0 GB
JE
Jebadiah 27b v2 1
Jebadiah · bartowski
Params27.3B
Context256K
Q4_K_M17.4 GB
LL
Humanish Roleplay Llama 3.1 8B i1
Llama · mradermacher
Params8B
Context128K
Best fitIQ2_XXS 4.0 GB
QW
Qwen3.8 4B Agentic Distill
Qwen · mradermacher
Params4.21B
Context256K
Best fitQ3_K_L 4.0 GB
FR
FrogNano 4B 2609
Frognano · mradermacher
Params4.33B
Context256K
Best fitQ3_K_M 3.9 GB
QW
Qwen3 VL Reranker 2B
Qwen · mradermacher
Params1.72B
Context256K
Best fitQ8_0 3.3 GB
TI
tinyllama_tinyllama-1.1b-intermediate-step-1431k-3t
Tinyllama · TheBloke
Params1.1B
Context2K
Best fitQ8_0 1.9 GB
MI
Mahou-1.2a-mistral-7B
Mistral · mradermacher
Params7.24B
Context32K
Best fitIQ2_S 3.9 GB
QW
Hana 0.1 Fast
Qwen · dwmk
Params8.03B
Context32K
Smallest0.3 GB