GGUF models that fit in 48 GB

201364 GGUF language models have at least one quant whose estimated memory use is 48 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
GE
Huihui Gemma 3n E4B It Abliterated
Gemma · bartowski
Params6.9B
Context32K
Best fitBF16 14.8 GB
GGUF
QW
Qwen3.5 9B Heretic
Qwen · mradermacher
Params8.95B
Context256K
Best fitF16 19.5 GB
GGUF
QW
Qwopus3.8 27B Flash V2 Apostate Uncensored
Qwopus · mradermacher
Params27.3B
Context256K
Best fitQ8_0 31.7 GB
Vision
MO
MolOptAgent 7B
Moloptagent · mradermacher
Params7.62B
Context32K
Best fitF16 16.2 GB
GGUF
YI
dolphin-2_2-yi-34b
Yi · mradermacher
Params34.4B
Context16K
Best fitQ6_K 30.7 GB
GGUF
XL
xLAM 2 3b Fc R
Xlam · Salesforce
Params3.09B
Context32K
Best fitF16 7.0 GB
Tool use
KO
Koa Express 31B Qat
Koa · mradermacher
Params32.1B
Context256K
Best fitIQ3_M 47.7 GB
ReasoningTool use
MI
Mistral Small 4 119B Uncensored
Mistral · timteh673
Params119B
Context1024K
Best fitQ2_K 44.2 GB
GGUF
GL
GLM 4.7 Flash Claude Opus 4.5 High Reasoning Distill
GLM · TeichAI
Params30B
Context198K
Best fitQ8_0 33.2 GB
Reasoning
GR
Granite 20b Code Base R1.1
Granite · RichardErkhov
Params20B
Context8K
Best fitQ8_0 22.2 GB
CodingReasoning
WI
models--microsoft--WizardLM-2-8x22B
Wizardlm · MaziyarPanahi
Params141B
Context64K
Best fitIQ1_M 35.1 GB
GGUF
QW
Qwen2.5 Coder 1.5B Instruct
Qwen · unsloth
Params1.54B
Context128K
Best fitF16 3.8 GB
Coding
DE
DeepSeek R1 Qwen3 8B heretic
DeepSeek · saidutta69
Params8.19B
Context128K
Best fitF16 18.1 GB
ReasoningRoleplayTool use
GE
Gemma 4 26B A4B It Heretic
Gemma · nohurry
Params25.2B
Context256K
Best fitQ8_0 35.4 GB
CodingVision
LL
Llama 2 70B
Llama · TheBloke
Params69B
Context4K
Best fitQ4_K_M 44.6 GB
GGUF
WI
WizardCoder 15B V1.0 i1
Wizardcoder · mradermacher
Params15.8B
Context8K
Best fitQ6_K 13.8 GB
Coding
WI
Wicked Stuff 12B
Wicked · ReadyArt
Params11.9B
Context256K
Best fitQ8_0 26.1 GB
ReasoningRoleplayTool use
SE
SEX_ROLEPLAY.v4 3.2 1B
Sex · mradermacher
Params1.5B
Context128K
Best fitQ6_K 2.0 GB
RoleplayTool use
LL
Llama 3.3 70B Instruct ablated
Llama · bartowski
Params70.6B
Context128K
Best fitQ4_K_L 46.5 GB
GGUF
LL
llava v1.5 7b llamafile
Llama · mozilla-ai
Params6.74B
Context4K
Best fitF16 18.3 GB
Vision
GE
Gemma 4 E2B
Gemma · mradermacher
Params4.65B
Context128K
Best fitQ6_K 4.9 GB
GGUF
GE
Gemma 4 31B Abliterated Uncensored
Gemma · mradermacher
Params30.7B
Context256K
Best fitQ3_K_M 48.0 GB
Roleplay
QW
Kataguru Qwen3.8 27B Titan v3.2
Qwen · kataguru
Params26.9B
Context256K
Best fitIQ4_NL 18.4 GB
ReasoningTool useVision
MI
Mimo-Vl-7B-Rl
Mimo · unsloth
Params7.62B
Context125K
Best fitBF16 17.0 GB
Vision
QW
Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive
Qwen · Unrestricted
Params34.7B
Context256K
Best fitQ8_K 44.8 GB
Vision
FI
Fimbulvetr-11B-v2.1-16K
Fimbulvetr · mradermacher
Params10.7B
Context16K
Best fitQ6_K 10.9 GB
GGUF
MU
Muse-Glimmer-30B-Hermes-Agentic
Muse · vcruz305
Params27.9B
Context128K
Best fitQ8_0 30.5 GB
Tool useVision
FL
flint
Flint · terelabs
Params39.8B
Context256K
Best fitQ5_K_M 30.3 GB
ReasoningTool use
QW
Qwen3.8 Whittle MoE 27B A17.8B
Qwen · mradermacher
Params27B
Context256K
Best fitQ8_0 31.3 GB
GGUF
QW
Qwen3.8 27B Humanlike Chat 2.0
Qwen · mradermacher
Params27.3B
Context256K
Best fitQ6_K 25.1 GB
RoleplayTool use

Other memory sizes

Keep browsing