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.
HE
Hermes 4.3 36B Heretic
Hermes · mradermacher
Params36.2B
Context512K
Q4_K_M21.8 GB
LL
Llama 3.2 3B Instruct Abliterated
Llama · mradermacher
Params3.61B
Context128K
Best fitQ5_K_S 4.0 GB
RI
Ringg Router E2B
Ringg · bartowski
Params4.63B
Context128K
Best fitQ3_K_L 3.7 GB
LL
Llama-3.1-Nemotron-Nano-8B-V1
Llama · unsloth
Params8B
Context128K
Best fitIQ1_M 3.9 GB
ME
Med Advisor Conversation 4B
Med · mradermacher
Params4.41B
Context40K
Best fitQ3_K_M 4.0 GB
GP
GPT OSS Swallow 120B RL v0.1
gpt-oss · mradermacher
Params117B
Context128K
Q4_K_M87.9 GB
NX
Nxcode-CQ-7B-orpo
Nxcode · legraphista
Params7.25B
Context64K
Best fitIQ2_M 4.0 GB
QW
Huihui Qwen3.5 35B A3B Claude 4.6 Opus Abliterated
Qwen · mradermacher
Params34.7B
Context256K
Q4_K_M21.2 GB
NE
Nemotron 3.5 Bolt 74B A9B CRACK 24GB
Nemotron · dealignai
Params74B
Context256K
Smallest1.0 GB
CO
Yi Coder 1.5B Chat
Coder · mradermacher
Params1.48B
Context128K
Best fitQ6_K 3.4 GB
BG
bge m3
BGE · CompendiumLabs
Params567M
Context8K
Best fitF32 3.6 GB
MI
Mahou-1.3-mistral-7B
Mistral · mradermacher
Params7.24B
Context32K
Best fitIQ2_S 3.9 GB
MI
MiniCPM V 4.7 1B
MiniCPM · mradermacher
Params1.01B
Context256K
Best fitQ6_K 1.8 GB
MI
Mistral Medium 3.5 128B
Mistral · bartowski
Params125B
Context256K
Q4_K_M78.4 GB
QW
Qwen3 VL 8B Instruct
Qwen · mradermacher
Params8.19B
Context256K
Best fitIQ1_M 4.0 GB
GE
Gemma 3 4b It Heretic Uncensored Abliterated Extreme
Gemma · mradermacher
Params3.88B
Context128K
Best fitQ4_K_S 4.0 GB
ST
stable-code-3b
Stable · mradermacher
Params2.8B
Context16K
Best fitIQ2_XXS 4.0 GB
LI
LightOnOCR 3 0.8B
Lightonocr · mradermacher
Params752M
Context256K
Best fitQ5_K_S 1.5 GB
HE
Hemmingway 1.5
Hemmingway · bartowski
Params27.3B
Context256K
Q4_K_M17.4 GB
OC
Occamy 1.0 APEX I NanoPlus
Occamy · IsValorum
Params34.7B
Context256K
Best fitQ8_0 3.2 GB
BE
BetterWright K2 Horizon 7B Uno
Betterwright · mradermacher
Params9B
Context512K
Q4_K_M5.6 GB
QW
Qwen2.5 Coder 3B Instruct
Qwen · unsloth
Params3.09B
Context32K
Best fitQ6_K 3.3 GB
MI
MiniCPM V 4.7 1B
MiniCPM · NANI-Nithin
Params1.01B
Context256K
Best fitBF16 2.9 GB
AV
Avrik Lite 4B
Avrik · mradermacher
Params7.52B
Context128K
Q4_K_M5.3 GB
GE
GELab Zero 4B Preview
Gelab · bartowski
Params4B
Context256K
Best fitIQ4_XS 4.0 GB
UN
Unlimited OCR 2
Unlimited · AtomicChat
Params2.94B
Context128K
Best fitQ6_K 3.7 GB
LL
rombos Llama3 8B Instruct Replete Adapted i1
Llama · mradermacher
Params8B
Context8K
Best fitIQ2_XXS 4.0 GB
QW
Qwen2.5 Coder 7B
Qwen · QuantFactory
Params7.6B
Context32K
Best fitQ2_K 4.0 GB
OR
Ornith 1.5 9B Heretic
Ornith · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
KE
Kev-4B Artemis 2.5GB
Kev · turintech
Params4.21B
Context256K
Best fitIQ3_M 3.7 GB