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
LL
dolphin 2.9.4 llama3.1 8b
Llama · bartowski
Params8B
Context128K
Best fitQ8_0 10.1 GB
Coding
RW
Rwkv 6 World 7b
Rwkv · bartowski
Params7.64B
Context1024K
Best fitQ8_0 10.0 GB
GGUF
LF
LFM2.5 8B A1B Hermes Agentic Coder Abliterated v3
Lfm · DuoNeural
Params8.47B
Context125K
Best fitQ8_0 11.1 GB
CodingTool use
FI
FinanceGemma E4B
Financegemma · mradermacher
Params7.5B
Context128K
Best fitQ8_0 9.9 GB
Vision
VE
Veloxy Flash 15B
Veloxy · mradermacher
Params8.95B
Context256K
Best fitQ6_K 8.9 GB
CodingReasoningTool use
QW
qwen image 2.1
Qwen · gguf-org
Params338M
Context256K
Q4_K_M5.0 GB
Vision
FA
Falcon H1R 7B
Falcon · tiiuae
Params7.59B
Context256K
Best fitQ8_0 8.9 GB
GGUF
MI
MiniCPM V 4.7 35B A3B APEX I MiniPlus V2.1
MiniCPM · IsValorum
Params34.7B
Context256K
Best fitQ6_K 0.9 GB
Vision
YI
Yi-1.5-6B-Chat
Yi · bartowski
Params6.06B
Context4K
Best fitQ8_0 7.5 GB
GGUF
LL
llm4decompile 9b v2
Llm · mradermacher
Params8.83B
Context128K
Best fitQ8_0 10.7 GB
GGUF
MI
Ministral 3 8B Reasoning 2512
Mistral · mistralai
Params8.49B
Context256K
Best fitQ8_0 10.7 GB
ReasoningVision
G9
G9v3 3B
bartowski
Params3B
Context128K
Best fitBF16 6.9 GB
Tool use
TE
Tev 27B
Tev · mradermacher
Params27B
Context256K
Best fitIQ2_S 12.0 GB
GGUF
SM
Smollm3-3B-128K
SmolLM · unsloth
Params3.08B
Context128K
Best fitBF16 7.3 GB
GGUF
GE
Huihui Gemma 4 E4B It Abliterated
Gemma · mradermacher
Params7.5B
Context128K
Best fitQ8_0 9.9 GB
Vision
FL
flux2 klein
Flux · ChrisColeTech
Params9.08B
Context40K
Best fitQ8_0 11.7 GB
Vision
VI
Vinci MLE 123B 1.0
Vinci · mradermacher
Params125B
Context256K
Q4_K_M74.9 GB
CodingTool use
KA
KAT Coder V2.5 Dev APEX I MiniPlus V2.1
Kat · IsValorum
Params34.7B
Context256K
Best fitQ4_0 2.4 GB
CodingReasoningTool use
MI
Mistral Heretica 12B
Mistral · AnkitAI
Params12.2B
Context128K
Best fitQ6_K 11.9 GB
Roleplay
QW
Sakura FLUX.3 Image Klein TextEncoder Uncensored
Qwen · webmp3
Params4B
Context256K
Best fitQ8_0 6.0 GB
CodingVision
SZ
SZL Khipu 1.5B
Szl · SZLHOLDINGS
Params1.54B
Context32K
Best fitF16 3.8 GB
Tool use
LL
Llama 3.2 3B Instruct
Llama · DavidAU
Params3.21B
Context128K
Best fitQ6_K 4.1 GB
Roleplay
QW
Huihui Qwen3 Coder Next Abliterated
Qwen · mradermacher
Params79.7B
Context256K
Q4_K_M48.5 GB
Coding
JI
Jina Embeddings v5 Text Small Classification
Jina · jinaai
Params596M
Context32K
Best fitF16 2.6 GB
Embedding
OL
OLMoE 1B 7B 0924 Instruct
OLMo · mradermacher
Params6.9B
Context4K
Best fitQ6_K 7.3 GB
GGUF
QW
Qwen3.8-27B
Qwen · ahmed-amedo
Params27.3B
Context256K
Best fitIQ2_S 11.9 GB
Vision
CY
Cydonia 24B v4.3 Absolute Heresy
Cydonia · mradermacher
Params23.6B
Context128K
Best fitIQ3_XS 11.7 GB
GGUF
NO
normistral 7b warm
Normistral · ltg
Params7.25B
Context2K
Best fitQ8_0 9.3 GB
Coding
LL
Llama 3.1 8B Lexi Uncensored V2
Llama · Orenguteng
Params8B
Context128K
Smallest4.7 GB
GGUF
QW
Qwen3-Vl-30B-A3B-Thinking
Qwen · unsloth
Params30.5B
Context256K
Best fitIQ2_XXS 11.6 GB
ReasoningVision

Other memory sizes

Keep browsing