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.

Most downloadedNewestLargestLongest contextSmallest
SA
Sakura SOLAR Instruct
Sakura · TheBloke
Params10.7B
Context4K
Best fitQ6_K 10.9 GB
GGUF
LL
Llama3.1 8B ShiningValiant2
Llama · bartowski
Params8B
Context128K
Best fitQ8_0 10.1 GB
GGUF
MI
MiniCPM V 4.7 1B
MiniCPM · NANI-Nithin
Params1.01B
Context256K
Best fitBF16 2.9 GB
Vision
AV
Avrik Lite 4B
Avrik · mradermacher
Params7.52B
Context128K
Best fitQ6_K 8.1 GB
GGUF
GE
GELab Zero 4B Preview
Gelab · bartowski
Params4B
Context256K
Best fitBF16 9.8 GB
Vision
MA
Magicoder S DS 6.7B
Magicoder · TheBloke
Params6.74B
Context16K
Best fitQ8_0 12.0 GB
Coding
QW
MN 2407 DSK QwQify v0.1 12B
Qwen · bartowski
Params12.2B
Context128K
Best fitQ6_K 11.9 GB
Reasoning
CY
Cydonia 24B v4.3 Absolute Heresy
Cydonia · mradermacher
Params23.6B
Context128K
Best fitQ2_K 10.7 GB
GGUF
QW
Qwen3.5 9b Claude 4.8 Opus reasoning
Qwen · TheCidSama
Params9.2B
Context256K
Smallest5.8 GB
CodingReasoning
DO
weyaxi_dolphin2.1-openorca-7b
Dolphin · TheBloke
Params7.24B
Context32K
Best fitQ8_0 9.3 GB
GGUF
LL
Llama-2-7b-WikiChat-fused
Llama · RichardErkhov
Params6.74B
Context4K
Best fitQ8_0 12.0 GB
GGUF
LL
CodeLlama 7b Instruct Hf
Llama · mradermacher
Params6.74B
Context16K
Best fitQ6_K 10.3 GB
Coding
QW
Video Companions
Qwen · off-grid-ai
Params4.02B
Context256K
Best fitQ4_K_M 5.4 GB
CodingTool useVision
ST
L3 Stheno Maid Blackroot Grand HORROR 16B
Stheno · DavidAU
Params16.5B
Context8K
Best fitIQ4_XS 11.9 GB
Roleplay
LL
Llama3Ternary 10b
Llama · CMSManhattan
Params9.64B
Context8K
Best fitTQ2_0 5.9 GB
GGUF
UN
Unlimited OCR 2
Unlimited · AtomicChat
Params2.94B
Context128K
Best fitBF16 6.9 GB
Vision
PH
Phi 4 Reasoning
Phi · bartowski
Params14.7B
Context32K
Best fitQ4_K_L 11.6 GB
CodingReasoning
LL
rombos Llama3 8B Instruct Replete Adapted i1
Llama · mradermacher
Params8B
Context8K
Best fitQ6_K 8.2 GB
CodingTool use
QW
Qwen2.5 Coder 7B
Qwen · QuantFactory
Params7.6B
Context32K
Best fitQ8_0 9.1 GB
Coding
MI
nondzu_mistral-7b-code-16k-qlora
Mistral · TheBloke
Params7.24B
Context32K
Best fitQ8_0 9.3 GB
Coding
YI
Faro Yi 34B i1
Yi · mradermacher
Params34.4B
Context195K
Best fitIQ2_XXS 11.8 GB
GGUF
WI
WizardCoder Python 13B V1.0 i1
Wizardcoder · mradermacher
Params13B
Context16K
Best fitIQ2_M 11.7 GB
Coding
OR
Ornith 1.5 9B Heretic
Ornith · mradermacher
Params8.95B
Context256K
Best fitQ6_K 8.9 GB
GGUF
CH
Chocolatine 14B Instruct DPO v1.2 i1
Chocolatine · mradermacher
Params14B
Context4K
Best fitQ5_K_S 11.8 GB
GGUF
KE
Kev-4B Artemis 2.5GB
Kev · turintech
Params4.21B
Context256K
Best fitQ4_K_M 4.1 GB
ReasoningTool use
IN
INTELLECT 3
Intellect · bartowski
Params107B
Context128K
Q4_K_M71.4 GB
Reasoning
EM
embeddinggemma-2
Embeddinggemma · AtomicChat
Params271M
Context256K
Best fitBF16 2.7 GB
EmbeddingVision
EM
Embeddinggemma 300M Qat Q4_0 Unquantized
Embeddinggemma · ggml-org
Params303M
Context2K
Best fitQ4_0 1.0 GB
Embedding
QW
Qwen3.5 4B Heretic 2xMerge DavidAU
Qwen · mradermacher
Params4.21B
Context256K
Best fitQ6_K 5.1 GB
Reasoning
LL
LosslessMegaCoder Llama2 7B Mini
Llama · TheBloke
Params6.74B
Context4K
Best fitQ8_0 12.0 GB
Coding

Other memory sizes

Keep browsing