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.
QW
Qwen2.5 VL 7B Abliterated Caption It
Qwen · mradermacher
Params7.62B
Context125K
Best fitQ6_K 7.2 GB
QW
Qwen3.5 4B Claude 4.6 OS Auto Variable HERETIC UNCENSORED THINKING
Qwen · mradermacher
Params4.21B
Context256K
Best fitF16 10.0 GB
LL
Llama 3.1 8B Lexi Uncensored
Llama · bartowski
Params8B
Context128K
Best fitQ8_0 10.1 GB
QW
Andy 4.2
Qwen · Mindcraft-CE
Params8.95B
Context256K
Best fitQ8_0 11.1 GB
MI
ehartford_dolphin-2.2.1-mistral-7b
Mistral · TheBloke
Params7.24B
Context32K
Best fitQ8_0 9.3 GB
QW
Qwen3 4B Abliterated
Qwen · mradermacher
Params4B
Context40K
Best fitF16 9.8 GB
LL
Phind CodeLlama 34B v1
Llama · mradermacher
Params33.7B
Context16K
Best fitIQ2_XXS 11.1 GB
MI
MiMo V2.6 Flash RL NVFP4
Mimo · ProCreations
Params310B
Context1024K
Best fitBF16 4.9 GB
MI
MiniCPM V 4.7 35B A3B Abliterated
MiniCPM · mradermacher
Params34.7B
Context256K
Best fitIQ2_S 11.8 GB
QW
MiMo-V2.6-Distill-Qwen-9B
Qwen · ggml-org
Params8.95B
Context256K
Best fitQ8_0 11.1 GB
QW
Qwen3 14B Abliterated
Qwen · AIgotahole
Params14.8B
Context40K
Best fitQ4_K_S 10.4 GB
LL
Llama3.1-8B-Chinese-Chat
Llama · shenzhi-wang
Params8B
Context128K
Best fitQ8_0 10.1 GB
DE
DeepSeek R1 0528 Qwen3 8B
DeepSeek · bartowski
Params8.19B
Context128K
Best fitQ8_0 10.4 GB
SP
rizzo flow
Spark · rizzoaiacademy
Params1.71B
Context1024K
Best fitBF16 9.9 GB
CO
Yi Coder 9B
Coder · RichardErkhov
Params8.83B
Context128K
Best fitQ8_0 10.7 GB
ER
Ernie-4.5-21B-A3B-Thinking
Ernie · unsloth
Params21.8B
Context128K
Best fitQ3_K_M 11.5 GB
QW
Huihui Qwen3.5 9B Claude 4.6 Opus Abliterated
Qwen · mradermacher
Params8.95B
Context256K
Best fitQ8_0 11.1 GB
QW
Qwen3.5 27B Abliterated
Qwen · mradermacher
Params27B
Context256K
Best fitIQ2_S 12.0 GB
DO
Dots3 Note Prev
Dots · bartowski
Params280B
Context512K
Q4_K_M171.1 GB
LL
Llama 3.2 1B Instruct
Llama · hugging-quants
Params1.24B
Context128K
Best fitQ8_0 2.1 GB
DE
DeepHat V1 7B
Deephat · mradermacher
Params7.62B
Context32K
Best fitQ8_0 9.1 GB
RO
RoMistral-7b-Instruct
Romistral · legraphista
Params7.24B
Context32K
Best fitQ8_0 9.3 GB
QW
Dongfangshuo Qwen3.8 27B NSFW
Qwen · mradermacher
Params27.3B
Context256K
Q4_K_M16.8 GB
YI
brucethemoose_yi-34b-200k-dare-megamerge-v8
Yi · TheBloke
Params34.4B
Context195K
Best fitIQ2_XXS 11.8 GB
YI
Yi 6B Chat
Yi · RichardErkhov
Params6.06B
Context4K
Best fitQ8_0 7.5 GB
MI
Abliterated Dolphin3.0 R1 Mistral 24B
Mistral · mradermacher
Params23.6B
Context32K
Best fitIQ3_XS 11.7 GB
QW
Qwen3.5 9B The Defiant Fable DARK ROAST Uncensored Heretic NEO IMATRIX MAX MTP
Qwen · mradermacher
Params9.2B
Context256K
Best fitQ6_K 9.2 GB
LF
LFM2.5 Embedding 350M
Lfm · LiquidAI
Params354M
Context125K
Best fitBF16 1.7 GB
QW
shisa v2.1 qwen3 8b
Qwen · mradermacher
Params8.19B
Context40K
Best fitQ8_0 10.4 GB
JA
Jan-Nano-128K
Jan · unsloth
Params4B
Context128K
Best fitBF16 9.8 GB