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.
ME
G4 MeroMero 26B A4B It Uncensored Heretic
Meromero · mradermacher
Params25.2B
Context256K
Best fitQ6_K 31.2 GB
FL
Flint-40B-A2.5B-abliterated
Flint · dima6312
Params39.8B
Context256K
Best fitQ8_0 43.7 GB
QW
Qwen3 128k 30B A3B NEO MAX Imatrix
Qwen · DavidAU
Params30.5B
Context128K
Best fitQ8_0 34.1 GB
QW
Qwen3 14B Abliterated
Qwen · mradermacher
Params14.8B
Context40K
Best fitQ8_0 17.5 GB
HO
K2 Horizon 3.7B
Horizon · IFM
Params5.06B
Context512K
Best fitBF16 11.8 GB
QW
Huihui Qwen3.5 35B A3B Abliterated
Qwen · mradermacher
Params34.7B
Context256K
Best fitQ6_K 29.7 GB
QW
Therapy 3.8
Qwen · Verdugie
Params26.9B
Context256K
Best fitQ8_0 31.2 GB
LI
Ling 3.0 Flash VL
Ling · bartowski
Params124B
Context128K
Best fitIQ1_M 47.3 GB
LF
LFM2.5-350M
Lfm · weareapexcreators
Params354M
Context125K
Best fitQ8_0 1.4 GB
GI
GigaChat3.1 10B A1.8B
Gigachat · ai-sage
Params10.7B
Context256K
Best fitBF16 22.3 GB
KI
Kitchoon V3 26B A4B
Kitchoon · mradermacher
Params25.2B
Context256K
Best fitQ6_K 31.2 GB
QW
Qwen3 VL 8B Instruct Abliterated v2
Qwen · prithivMLmods
Params8.19B
Context256K
Best fitF16 18.1 GB
QW
Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled v2
Qwen · Jackrong
Params27B
Context256K
Best fitQ8_0 31.2 GB
JE
JEV 27B VL
Jev · Atlas3D
Params27B
Context256K
Best fitQ8_0 31.2 GB
ST
StartLux-Decision-v1.1-4B-Instant
Startlux · startlux-models
Params4.2B
Context256K
Best fitQ8_0 6.1 GB
QW
Qwen3 30B A3B
Qwen · bartowski
Params30.5B
Context32K
Best fitQ8_0 33.8 GB
ST
StartLux-Decision-v1.1-27B-Instant
Startlux · startlux-models
Params27B
Context256K
Best fitQ8_0 31.2 GB
QW
Qwen3.8-27B-DFlash2
Qwen · perplexity-ai
Params27.3B
Context256K
Best fitQ4_K_M 17.5 GB
FL
vllm-d3-flash
Flash · ggml-org
Params8.95B
Context256K
Best fitBF16 19.5 GB
QW
Qwen3.8 9B Distill uncensored heretic Q4 K M
Qwen · ge525
Params9.2B
Context256K
Best fitQ4_K_M 7.4 GB
GE
Gemma 3 12b It
Gemma · bartowski
Params11.8B
Context128K
Best fitBF16 27.3 GB
FL
RecursiveSI flint
Flint · bartowski
Params39.8B
Context256K
Best fitQ8_0 43.7 GB
GE
Sprinkle Gemma 4 31B
Gemma · mradermacher
Params30.7B
Context256K
Best fitQ3_K_M 48.0 GB
LF
LFM2 VL 1.6B
Lfm · LiquidAI
Params1.17B
Context125K
Best fitF16 3.9 GB
ST
L3-8B-Stheno-v3.2
Stheno · Lewdiculous
Params8B
Context8K
Best fitQ8_0 10.1 GB
LL
VAGOsolutions Llama 3 SauerkrautLM 70b Instruct
Llama · RichardErkhov
Params6B
Context—
Best fitIQ3_XS 33.3 GB
PA
PaddleOCR VL 1.5
Paddleocr · PaddlePaddle
Params467M
Context128K
Smallest0.9 GB
QW
Qwen3.5 9B Claude 4.6 Opus Reasoning Distilled
Qwen · Jackrong
Params8.95B
Context256K
Best fitQ8_0 11.1 GB
GE
gemma-2b-it
Gemma · lmstudio-ai
Params2.51B
Context8K
Best fitQ8_0 3.3 GB
QW
Atomic Chat / Qwen3.5 9B
Qwen · AtomicChat
Params8.95B
Context256K
Best fitQ8_0 11.1 GB