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.
ON
OneJev 0.8B
Onejev · mradermacher
Params1.01B
Context256K
Best fitF16 2.9 GB
MI
OpenChat 3.5 7B Mixtral v2.0 i1
Mistral · mradermacher
Params7.24B
Context8K
Best fitIQ2_S 3.9 GB
MI
MiniCPM5-2B
MiniCPM · Mungert
Params2.52B
Context128K
Best fitQ8_0 3.5 GB
QW
Qwen3.5-2B
Qwen · trubisky
Params1.88B
Context256K
Best fitIQ3_M 1.9 GB
AR
arco reasoner i1
Arco · mradermacher
Params514M
Context8K
Best fitQ5_K_S 1.3 GB
DE
Decka 4B
Decka · mradermacher
Params4.21B
Context256K
Best fitQ3_K_L 4.0 GB
QW
Qwen3 14B RefusalDirection ThinkingAware
Qwen · mradermacher
Params14.8B
Context40K
Q4_K_M9.0 GB
GE
Gemma 3 270m It
Gemma · gguf-org
Params268M
Context32K
Best fitF32 1.7 GB
QW
GPT 5 Distill Qwen3 4B Instruct
Qwen · Jackrong
Params4B
Context256K
Best fitIQ4_XS 4.0 GB
LL
Impish_LLAMA_3B
Llama · l3utterfly
Params3.2B
Context128K
Best fitQ4_K_M 3.5 GB
QW
Qwen3.5 4B Heretic
Qwen · mradermacher
Params4.21B
Context256K
Best fitIQ4_XS 4.0 GB
NA
NaNovel 9B
Nanovel · mradermacher
Params8.95B
Context256K
Best fitIQ1_S 4.0 GB
US
UserRepGen 4B
Userrepgen · bartowski
Params4.21B
Context256K
Best fitIQ4_XS 4.0 GB
OM
OmniCoder 9B
Omnicoder · bartowski
Params8.95B
Context256K
Q4_K_M5.9 GB
HU
HuatuoGPT 3 9B
Huatuogpt · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
QW
Qwopus3 5 9B Coder Flash
Qwopus · mradermacher
Params9.2B
Context256K
Q4_K_M5.8 GB
QW
Qwen3 VL 8B Thinking
Qwen · bartowski
Params8.19B
Context256K
Q4_K_M5.0 GB
QW
Qwen2.5 0.5B Instruct
Qwen · mradermacher
Params494M
Context32K
Q4_K_M0.4 GB
RU
RusMind.Моментум 3.5.1
Rusmind · DICPlay
Params4.65B
Context128K
Q4_K_M3.4 GB
LR
OrionLLM LRM 3.2
Lrm · bartowski
Params27B
Context256K
Q4_K_M17.5 GB
LL
Herplete LLM Llama 3.1 8b i1
Llama · mradermacher
Params8B
Context128K
Best fitIQ2_XXS 4.0 GB
SM
SmolLlamix 8x101M take2
Smolllamix · mradermacher
Params399M
Context32K
Best fitQ8_0 1.0 GB
GI
GigaChat3 10B A1.8B
Gigachat · bartowski
Params10.7B
Context256K
Q4_K_M6.7 GB
QW
Qwen2.5 VL 3B uncensored bughunter v9
Qwen · mmp2055
Params3.1B
Context125K
Best fitQ4_K_M 2.7 GB
AI
Aitana 7B S Instruct
Aitana · mradermacher
Params7.77B
Context8K
Best fitIQ1_M 4.0 GB
QW
MiMo V2.6 Distill Qwen 9B Ablit
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
QW
open text detector qwen36 35b a3b oct10
Qwen · woog
Params34.7B
Context256K
Q4_K_M21.2 GB
GP
Gpt2 Large
Gpt · mradermacher
Params838M
Context1K
Best fitF16 3.7 GB
FA
Fara 7B
Fara · gguf-org
Params7.62B
Context125K
Best fitQ2_K 4.0 GB
AP
Apocrypha-7b
Apocrypha · mradermacher
Params7.24B
Context32K
Best fitIQ2_S 3.9 GB