GGUF models that fit in 12 GB
171517 GGUF language models have at least one quant whose estimated memory use is 12 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 4194304K tokens.
ME
Merousis 26B A4B
Merousis · mradermacher
Params26B
Context256K
Q4_K_M17.2 GB
QW
Qwen3 Swallow 8B RL v0.2
Qwen · mmnga-o
Params8.19B
Context40K
Best fitQ8_0 10.4 GB
ME
Medgemma-27B-It
Medgemma · unsloth
Params27B
Context128K
Best fitIQ1_M 11.6 GB
AN
Anubis 70B v1.2
Anubis · bartowski
Params70.6B
Context128K
Q4_K_M42.5 GB
QW
Huihui Qwen3 Next 80B A3B Instruct Abliterated
Qwen · mradermacher
Params80B
Context256K
Q4_K_M48.4 GB
QW
sakura-8b-qwen3-v1.1
Qwen · SakuraLLM
Params8.2B
Context32K
Smallest4.6 GB
GL
glm-4-9b-chat
GLM · legraphista
Params9.4B
Context128K
Best fitQ8_0 10.8 GB
QW
ThinkingCap Qwen3.6 27B abliterated MTP
Qwen · protoLabsAI
Params27.3B
Context—
Best fitQ8_0 4.0 GB
FI
Finch 3b
Finch · latestissue
Params3.1B
Context1024K
Best fitBF16 7.5 GB
PH
Phi 3.1 mini 128k instruct
Phi · bartowski
Params3.82B
Context128K
Best fitQ8_0 7.8 GB
QW
Agnes 3.0 Qwen
Qwen · bartowski
Params180B
Context1024K
Q4_K_M127.1 GB
GE
Gemma-2-9B-It-SPPO-Iter3
Gemma · legraphista
Params9.24B
Context8K
Best fitQ6_K 10.9 GB
KI
Kimi VL A3B Instruct
Kimi · mradermacher
Params16B
Context128K
Best fitQ4_K_M 11.5 GB
LI
LightOnOCR 3 1B
Lightonocr · prithivMLmods
Params596M
Context16K
Best fitBF16 2.6 GB
MI
MiLMMT 46 12B v1.0
Milmmt · mradermacher
Params11.8B
Context128K
Best fitQ5_K_S 12.0 GB
LL
Llama 3.1 Swallow 8B Instruct v0.5
Llama · mmnga
Params8B
Context128K
Best fitQ8_0 10.1 GB
QW
Qwen2.5 Vl 7b Cam Motion
Qwen · mradermacher
Params7.62B
Context125K
Best fitQ8_0 9.1 GB
DE
DeepSeek R1 Distill Llama 8B Heretic
DeepSeek · saidutta69
Params8.03B
Context128K
Best fitQ8_0 10.1 GB
QW
Qwen3.8-27B
Qwen · jj33512
Params27.3B
Context256K
Best fitIQ2_S 11.9 GB
QW
Irwan LightOnOCR 3 0.8B Abliterated
Qwen · irwanworld
Params752M
Context256K
Best fitQ8_0 1.7 GB
OL
Olmo-3-7B-Instruct
OLMo · unsloth
Params7.3B
Context64K
Best fitQ6_K_XL 11.6 GB
KI
Kitchoon 26B A4B
Kitchoon · mradermacher
Params26B
Context256K
Q4_K_M17.2 GB
QW
Qwen3 4B Instruct 2507 Heretic
Qwen · bartowski
Params4B
Context256K
Best fitBF16 9.8 GB
MI
Ministral 3 14B Instruct 2512
Mistral · bartowski
Params13.5B
Context256K
Best fitQ5_K_L 11.9 GB
PH
dolphin-2.9.2-Phi-3-Medium
Phi · legraphista
Params14B
Context4K
Best fitQ5_K_S 11.8 GB
GR
Granite-4.0-350M
Granite · unsloth
Params352M
Context32K
Best fitBF16 2.1 GB
NA
Nanbeige4.2 3B
Nanbeige · owao
Params4.17B
Context256K
Best fitQ8_0 5.7 GB
WI
Wicked Stuff 26B A4B Final
Wicked · mradermacher
Params25.2B
Context256K
Q4_K_M16.8 GB
GL
GLM 4.5 Air Derestricted
GLM · bartowski
Params110B
Context128K
Q4_K_M73.5 GB
QW
Qwen2.5 Coder 3B Instruct
Qwen · lmstudio-community
Params3.09B
Context32K
Best fitQ8_0 4.1 GB