GGUF models that fit in 32 GB

198257 GGUF language models have at least one quant whose estimated memory use is 32 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
OR
orpheus 3b 0.1 ft Q4 K M
Orpheus · isaiahbjork
Params3.78B
Context128K
Best fitQ4_K_M 3.8 GB
GGUF
AI
Airoboros L2 13B 2.1
Airoboros · TheBloke
Params13B
Context4K
Best fitQ8_0 21.0 GB
GGUF
AP
Apex-35B-V1.0
Apex · Apexintelligence-AI
Params35.5B
Context256K
Best fitQ6_K_S 31.0 GB
ReasoningTool use
NE
NeuralLemon 7B
Neurallemon · backyardai
Params7.24B
Context32K
Best fitF32 30.5 GB
GGUF
GE
ThinkingCap Gemma 4 26B A4B It
Gemma · mradermacher
Params25.2B
Context256K
Best fitQ6_K 31.2 GB
ReasoningTool use
QW
FLUX 3 Image Klein TE Uncensored
Qwen · AtomicChat
Params4.02B
Context256K
Best fitBF16 9.8 GB
CodingTool useVision
LL
tachyphylaxis Meta Llama 3.1 Chat Uncensored
Llama · RichardErkhov
Params15M
Context—
Best fitQ2_K 30.0 GB
GGUF
DE
DeepSeek-V2-Lite-Chat
DeepSeek · mradermacher
Params15.7B
Context160K
Best fitQ8_0 19.5 GB
GGUF
QW
Qwen3 VL 4B Instruct C_Abliterated v2
Qwen · mradermacher
Params4B
Context256K
Best fitF16 9.8 GB
CodingVision
OR
Ornith 1.5 35B A3B
Ornith · ArtomYuan
Params35.5B
Context256K
Best fitQ6_0 30.6 GB
Vision
QW
Satyr V0.1 4B
Qwen · PantheonUnbound
Params4B
Context256K
Best fitF16 9.8 GB
Reasoning
LL
Llm Jp 4.1 33b Thinking
Llm · llm-jp
Params33.2B
Context64K
Best fitQ4_K_M 23.1 GB
Reasoning
QU
Quyet 1.0 Medium
Quyet · mradermacher
Params4.33B
Context256K
Best fitF16 10.3 GB
Vision
LL
KISTI KONI KONI Llama3 8B Instruct 20240729
Llama · RichardErkhov
Params8B
Context—
Best fitQ2_K 2.7 GB
GGUF
NO
2B_or_not_2B
Not · mradermacher
Params2.5B
Context8K
Best fitQ6_K 2.7 GB
GGUF
OR
Ornith 1.5 35B A3B Uncensored
Ornith · mradermacher
Params35.5B
Context256K
Best fitQ6_K 30.4 GB
Vision
QW
Dolphin3.0 Qwen2.5 1.5B
Qwen · bartowski
Params1.54B
Context128K
Best fitF32 6.9 GB
CodingTool use
LL
Llama 2 13B chat
Llama · TheBloke
Params13B
Context4K
Best fitQ8_0 21.0 GB
GGUF
QW
Qwen3-Vl-4B-Thinking
Qwen · unsloth
Params4B
Context256K
Best fitBF16 9.8 GB
ReasoningVision
QW
Qwen3.8 27B Uncensored FP8
Qwen · theresa00l
Params27.3B
Context256K
Best fitQ4_K_M 19.5 GB
ReasoningTool useVision
LL
Llama 3.1 8B Lexi Uncensored V2
Llama · bartowski
Params8B
Context128K
Best fitQ8_0 10.1 GB
GGUF
QW
Qwen3.8 4B Distilled
Qwen · Ma7ee7
Params4B
Context256K
Best fitQ8_0 6.0 GB
Reasoning
NE
NVIDIA-Nemotron-3-Nano-30B-A3B
Nemotron · ggml-org
Params31.6B
Context1024K
Best fitQ4_K_M 29.9 GB
GGUF
OL
olmOCR 2 7B 1025
OLMo · bartowski
Params7.62B
Context125K
Best fitBF16 16.2 GB
Vision
SA
Sabi Noema
Sabi · mradermacher
Params26.9B
Context256K
Best fitQ8_0 31.2 GB
ReasoningTool useVision
QW
Qwen3 235B A22B Abliterated
Qwen · mradermacher
Params235B
Context40K
Q4_K_M142.2 GB
GGUF
QW
Qwen3.5 9B Uncensored HauhauCS Aggressive Q4 K M
Qwen · Manojb
Params8.95B
Context256K
Best fitQ4_K_M 7.2 GB
Vision
LA
Laguna S 2.1 Uncensored APEX
Laguna · SC117
Params118B
Context1024K
Best fitBF16 2.9 GB
GGUF
JE
Jev-Style-0.8B-Decision-v3
Jev · chaoliangUNSW
Params752M
Context256K
Best fitF16 2.4 GB
ReasoningTool use
QW
RuadaptQwen3 4B Instruct
Qwen · RefalMachine
Params4B
Context256K
Best fitBF16 9.7 GB
GGUF

Other memory sizes

Keep browsing