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.
GE
Gemma-3-27B-It-Qat
Gemma · unsloth
Params27B
Context128K
Best fitIQ1_M 11.6 GB
LF
LFM2.5 24B A2B
Lfm · bartowski
Params24B
Context128K
Best fitIQ2_M 12.0 GB
EU
EuroLLM 22B Instruct 2512
Eurollm · bartowski
Params22.6B
Context32K
Best fitIQ3_XS 11.9 GB
FI
Finch 7b
Finch · latestissue
Params7.64B
Context1024K
Best fitQ8_0 9.9 GB
MU
Muse-Glimmer-30B
Muse · ggml-org
Params27.9B
Context128K
Best fitBF16 6.1 GB
QW
Qwen3.8 27B Uncensored
Qwen · MiawTeam
Params27.3B
Context256K
Best fitQ8_0 5.8 GB
QW
Qwen3 VL 8B NSFW Caption V4.5 Heretic
Qwen · mradermacher
Params8.19B
Context256K
Best fitQ8_0 10.4 GB
PH
phonon 40m en
Phonon · gradium
Params51M
Context—
Smallest0.1 GB
GE
Writer
Gemma · mradermacher
Params25.2B
Context256K
Q4_K_M16.8 GB
AE
AesCode 8B
Aescode · bartowski
Params8.19B
Context256K
Best fitQ8_0 10.4 GB
IM
imgmodel
Imgmodel · koboldcpp
Params3.47B
Context40K
Best fitQ5_1 8.0 GB
ST
Dolphincoder Starcoder2 15b
StarCoder · mradermacher
Params16B
Context16K
Best fitQ4_K_M 11.0 GB
LL
MiniCPM Llama3 V 2 5
Llama · openbmb
Params8.2B
Context8K
Best fitQ8_0 10.1 GB
OV
OvisOCR3
Ovisocr · mradermacher
Params752M
Context256K
Best fitF16 2.4 GB
DE
DeepSeek R1 Distill Llama 8B
DeepSeek · lmstudio-community
Params8B
Context128K
Best fitQ8_0 10.1 GB
RO
RoLlama2-7b-Base
Rollama · legraphista
Params6.74B
Context4K
Best fitQ8_0 12.0 GB
ME
Medgemma 4b It
Medgemma · bartowski
Params3.88B
Context128K
Best fitBF16 9.4 GB
LF
LFM2.5 2.6B Heretic Abliterated
Lfm · Abiray
Params2.7B
Context125K
Best fitQ8_0 5.4 GB
BA
Bathory X 26B A4B 0.2
Bathory · mradermacher
Params26B
Context256K
Q4_K_M17.2 GB
NE
nexus finance
Nexus · King3Djbl
Params1.54B
Context32K
Best fitF16 3.8 GB
LL
FILM6912 Meta Llama 3.1 8B Instruct
Llama · RichardErkhov
Params8B
Context—
Best fitQ3_K 5.0 GB
GR
Granite 3.3 8b Instruct
Granite · ibm-granite
Params8.17B
Context128K
Best fitQ8_0 10.5 GB
WE
WestLake 10.7B v2
Westlake · backyardai
Params10.7B
Context32K
Best fitQ6_K 10.9 GB
MI
Mistral Small 24B Instruct 2501
Mistral · unsloth
Params23.6B
Context32K
Best fitQ2_K_L 10.9 GB
LL
Llama Artemis
Llama · mradermacher
Params8.2B
Context128K
Best fitQ6_K 8.3 GB
MI
MiniCPM V 4
MiniCPM · openbmb
Params3.6B
Context32K
Best fitQ8_0 4.6 GB
DE
Qwen3.5 9B Distill Deepseek V4.1 Flash 20261004
DeepSeek · kevinzhou21
Params9.2B
Context256K
Best fitQ8_0 11.4 GB
FL
Flux2 Klein 9b Uncensored Text Encoder
Flux · WANG-CC
Params8.19B
Context40K
Best fitQ8_0 10.4 GB
IN
Ingot VII 27B Q2
Ingot · dendriteholdings
Params26.9B
Context256K
Smallest7.2 GB
QW
Qwen2.5 7B Instruct Abliterated v2
Qwen · lumen-models
Params7.62B
Context32K
Best fitQ4_K_M 5.7 GB