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.
GR
Granite 20b Code Instruct
Granite · bartowski
Params20B
Context8K
Best fitF16 41.0 GB
HO
Holo4 35B A3B
Holo · Hcompany
Params34.7B
Context256K
Best fitQ4_K_M 22.5 GB
GE
Fallen Gemma3 27B v1
Gemma · bartowski
Params27B
Context128K
Best fitQ8_0 33.4 GB
FO
Foundation Sec 8B Reasoning
Foundation · fdtn-ai
Params8B
Context128K
Best fitQ4_K_M 6.5 GB
SA
Sabi Noema
Sabi · owao
Params27B
Context256K
Best fitQ8_0 31.2 GB
QW
qwen2.5-32b-instruct
Qwen · Qwen
Params32.8B
Context128K
Best fitQ8_0 37.5 GB
HO
K2-Horizon-MoVA-36B-A4B
Horizon · mradermacher
Params37.4B
Context512K
Best fitQ6_K 32.9 GB
QW
Qwen3 8B Heretic
Qwen · mradermacher
Params8.19B
Context40K
Best fitF16 18.1 GB
QW
Qwen3.6 35B A3B Uncensored Genesis Hermes Final NVFP4
Qwen · jan1k
Params34.7B
Context256K
Smallest21.1 GB
QW
Qwen3.5 9B Coder
Qwen · mradermacher
Params9.2B
Context256K
Best fitF16 20.0 GB
MI
Mistral Small 3.1 24B Instruct 2503
Mistral · bartowski
Params23.6B
Context128K
Best fitQ8_0 26.9 GB
MI
MiniCPM5 2B Math
MiniCPM · Caldalis
Params2.52B
Context128K
Best fitBF16 5.9 GB
QW
Qwen3.6 35B Uncensored HauhauCS 1M
Qwen · satgeze
Params447M
Context1024K
Best fitQ4_K_M 22.9 GB
MI
Ministral-3-14B-Reasoning-2512
Mistral · unsloth
Params13.5B
Context256K
Best fitBF16 28.9 GB
NE
NVIDIA Nemotron 3.5 Bolt E4B Uncensored
Nemotron · AtomicChat
Params7.5B
Context64K
Best fitQ6_K 9.3 GB
ST
StartLux Decision 35B A3B
Startlux · mradermacher
Params35.5B
Context256K
Best fitQ8_0 39.0 GB
MA
UI Mate 9B
Mate · bartowski
Params8.95B
Context256K
Best fitBF16 19.5 GB
ST
StartLux Decision v1 27B Q4 K M
Startlux · startlux-models
Params27B
Context256K
Best fitQ4_K_M 19.2 GB
QW
Qwen3.8-27B
Qwen · julianmb
Params27.3B
Context256K
Smallest9.2 GB
TE
Ternary Bonsai 2 27B Abliterated PQ2 0 MTP
Ternary · BoldingBuilds
Params27B
Context256K
Smallest7.2 GB
GE
Gemma 2 27b It
Gemma · legraphista
Params27.2B
Context8K
Best fitQ8_0 32.5 GB
DE
DeepSeek R1 Distill Llama 70B Abliterated
DeepSeek · bartowski
Params70.6B
Context128K
Best fitQ4_1 47.5 GB
FA
Fara1.5 9B
Fara · bartowski
Params8.95B
Context256K
Best fitBF16 19.5 GB
QW
Qwen3.8-27B-Uncensored-Aggressive
Qwen · 0xKitkat
Params27.3B
Context256K
Best fitQ6_K 30.1 GB
QW
Gliese Qwen3.5 9B Abliterated Caption
Qwen · prithivMLmods
Params8.95B
Context256K
Best fitF32 37.4 GB
MI
MiniCPM V 4.7 1B heretic
MiniCPM · saidutta69
Params1.01B
Context256K
Best fitF16 2.9 GB
QW
Qwen3.8 27B Uncensored IQ4 XS MTP 16GB VRAM
Qwen · Bucoid
Params27.3B
Context256K
Best fitIQ4_XS 16.6 GB
DA
DanTagGen beta
Dantaggen · KBlueLeaf
Params393M
Context512
Best fitF16 2.1 GB
QW
Qwen3 VL 30B A3B Instruct
Qwen · bartowski
Params30.5B
Context256K
Best fitQ8_0 33.8 GB
QW
Qwen2.5 14B_Uncencored
Qwen · bartowski
Params14.8B
Context128K
Best fitF16 31.7 GB