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.
GE
Gemma 4 31B StyleTune
Gemma · mradermacher
Params32.1B
Context256K
Best fitIQ3_M 47.7 GB
QW
Qwen2.5 14B
Qwen · mradermacher
Params14.8B
Context128K
Q4_K_M9.0 GB
GE
Gemma 3 27b It
Gemma · ggml-org
Params27B
Context128K
Best fitQ8_0 33.4 GB
IN
Ingot X 27B Q2
Ingot · dendriteholdings
Params26.9B
Context256K
Smallest7.2 GB
WI
Wicked Stuff 26B A4B v2.1 FINAL
Wicked · mradermacher
Params25.2B
Context256K
Best fitQ6_K 31.2 GB
YI
Yi 1.5 34B
Yi · RichardErkhov
Params34.4B
Context4K
Best fitQ8_0 39.1 GB
QW
Qwen3 VL 8B Instruct Abliterated v2
Qwen · qign
Params8.19B
Context256K
Best fitF16 18.1 GB
HO
Holo 3.1 4B
Holo · prithivMLmods
Params4.84B
Context256K
Best fitBF16 11.3 GB
SE
SexyGPT v3 Thinking Female
Sexygpt · mradermacher
Params27.3B
Context256K
Best fitQ6_K 25.1 GB
MT
Mtreelva Dune 12B
Mtreelva · mradermacher
Params12.9B
Context256K
Best fitQ6_K 24.0 GB
OU
OuteTTS 0.2 500M
Outetts · OuteAI
Params500M
Context32K
Q4_K_M0.4 GB
SA
SaulLM 54B Instruct i1
Saullm · mradermacher
Params46.7B
Context32K
Best fitQ6_K 40.0 GB
QW
Qwen3 VL 8B Instruct Heretic
Qwen · mradermacher
Params8.19B
Context256K
Best fitF16 18.1 GB
GP
Gpt Oss 20b Heretic Ara v3
gpt-oss · mradermacher
Params20.9B
Context128K
Best fitQ8_0 23.2 GB
QW
MiniMax H3 TextEncoder 16GB abliterated
Qwen · pottokao
Params25.8B
Context256K
Smallest12.6 GB
SE
SenseNova U1 8B MoT SFT
Sensenova · mradermacher
Params9.35B
Context256K
Best fitQ6_K 9.6 GB
QW
Himeros V2 27B
Qwen · Skttttt
Params27B
Context256K
Best fitQ8_0 31.2 GB
TI
Tifa Deepsex 14b CoT
Tifa · mradermacher
Params14.8B
Context128K
Best fitQ8_0 17.8 GB
GR
Kiesel v2
Granite · mradermacher
Params8.8B
Context128K
Best fitQ6_K 9.1 GB
LL
Akikawa 3.3 1b Roleplay 128K
Llama · Green-Eye
Params1.24B
Context128K
Best fitF16 3.2 GB
LL
bangla-llama-7b-instruct-v0.1
Llama · mradermacher
Params6.89B
Context4K
Best fitQ6_K 10.4 GB
HO
K2 Horizon 3.7B
Horizon · NANI-Nithin
Params5.06B
Context512K
Best fitBF16 11.8 GB
GR
Granite 20b Functioncalling
Granite · RichardErkhov
Params20B
Context8K
Best fitQ8_0 22.2 GB
KI
Kitchoon V3.1 26B A4B
Kitchoon · mradermacher
Params25.2B
Context256K
Best fitQ6_K 31.2 GB
QW
Eve V2 Unleashed Qwen3.5 8B Liberated 4K 4B Merged
Qwen · JeffGreen311
Params4.66B
Context256K
Best fitQ4_K_M 8.2 GB
TI
TimeCrystal l2 13B i1
Timecrystal · mradermacher
Params13B
Context4K
Best fitQ6_K 18.1 GB
PY
Pygmalion 2 13B
Pygmalion · TheBloke
Params13B
Context4K
Best fitQ8_0 21.0 GB
QW
Qwen3.5 4B
Qwen · gguf
Params4.33B
Context256K
Best fitQ8_0 6.2 GB
GR
Granite 4.0 H 1b
Granite · ibm-granite
Params1.46B
Context1024K
Best fitBF16 5.4 GB
TE
NaviDC OCR
Teleocr · nandraj
Params752M
Context125K
Best fitF16 2.9 GB