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.
JI
Jina Embeddings v5 Text Small Text Matching
Jina · jinaai
Params596M
Context32K
Best fitF16 2.6 GB
QW
Qwen3.5 27B Uncensored Heretic v1
Qwen · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
QW
Qwen3.5 4B Uncensored
Qwen · mradermacher
Params4.33B
Context256K
Best fitQ3_K_M 3.9 GB
QW
Qwen3.5 0.8B Claude 4.6 Opus Reasoning Distilled
Qwen · Jackrong
Params752M
Context256K
Best fitQ8_0 1.7 GB
PR
PreciseCoder 27B
Precisecoder · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
QW
Qwen2.5 0.5B Instruct
Qwen · ThomasBaruzier
Params494M
Context32K
Best fitF16 1.6 GB
MI
MiMo Negentropy 9B AGSI
Mimo · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
KR
Kronumos Kairos v2
Kronumos · bartowski
Params7.6B
Context32K
Best fitIQ2_M 3.9 GB
LL
Llama 3.3 70B Instruct Abliterated v2
Llama · mradermacher
Params70.6B
Context128K
Q4_K_M42.5 GB
PR
Prentis EQ 2
Prentis · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
HE
Hermes 4 70B
Hermes · bartowski
Params70.6B
Context128K
Q4_K_M42.5 GB
LL
Hermes-2-Theta-Llama-3-8B
Llama · legraphista
Params8B
Context8K
Best fitIQ2_XXS 4.0 GB
OR
Ornith 1.5 27B A3B Coder MTP
Ornith · ManniX-ITA
Params26.2B
Context256K
Q4_K_M16.1 GB
LL
Llama 3.3 8B Instruct
Llama · bartowski
Params8B
Context8K
Q4_K_M4.9 GB
PR
PreciseCoder 27B GRPO 0.5F1
Precisecoder · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
CU
CursorCore DS 1.3B SR i1
Cursorcore · mradermacher
Params1.35B
Context16K
Best fitQ6_K 3.3 GB
QW
DeepSeek V4 Pro Qwen3.5 4B MTP Public 20260801 HF
Qwen · Jackrong
Params4.33B
Context256K
Best fitQ3_K_M 3.9 GB
TE
Tev1 4B Experimental
Tev · bartowski
Params4.33B
Context256K
Best fitIQ3_M 4.0 GB
QW
Qwen2.5 VL 7B NSFW Caption V3 Abliterated
Qwen · mradermacher
Params7.62B
Context125K
Best fitQ2_K 4.0 GB
QW
Qwen3.5 2B Uncensored Zero Refusal SOMA
Qwen · AkaumiKaede
Params1.88B
Context256K
Best fitQ8_0 2.9 GB
LI
Ling Flash 2.0
Ling · bartowski
Params103B
Context32K
Q4_K_M62.6 GB
QW
Qwen3.5 9b Uncensored
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
RA
MS Raven 24B v1
Raven · mradermacher
Params23.6B
Context128K
Q4_K_M14.3 GB
GE
Gemma 3 12b It Vl Kimi V2 Heretic Uncensored Thinking
Gemma · mradermacher
Params11.8B
Context128K
Q4_K_M7.3 GB
LF
LFM3-1.7B
Lfm · AtomicChat
Params1.73B
Context256K
Best fitQ8_0 4.0 GB
WR
Wren
Wren · ohmysimo
Params117B
Context256K
Q4_K_M77.6 GB
QW
MiMo V2.6 Distill Qwen 9B Heretic
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
HU
Hunyuan 7B Instruct
Hunyuan · bartowski
Params7.5B
Context256K
Q4_K_M4.6 GB
QW
Huihui Qwen3.8 27B Abliterated
Qwen · mradermacher
Params27.3B
Context256K
Q4_K_M16.8 GB
GE
Gemma 3n E4B It
Gemma · bartowski
Params6.9B
Context32K
Best fitQ2_K 3.8 GB