GGUF models that fit in 4 GB

79024 GGUF language models have at least one quant whose estimated memory use is 4 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 1280000K tokens.

Most downloadedNewestLargestLongest contextSmallest
NA
Nanbeige4.2 3B
Nanbeige · bartowski
Params4.17B
Context256K
Best fitQ4_1 4.0 GB
GGUF
QW
Qwen3 ASR 0.6B
Qwen · ggml-org
Params752M
Context64K
Best fitBF16 2.9 GB
Vision
LF
LFM2.5 1.2B Instruct LocalAI
Lfm · weareapexcreators
Params1.17B
Context125K
Best fitQ8_0 2.8 GB
GGUF
QW
Atomic Chat / Qwen3.5 4B
Qwen · AtomicChat
Params4.21B
Context256K
Best fitQ3_K_L 4.0 GB
ReasoningTool use
QW
Qwen3 1.7B
Qwen · lmstudio-community
Params2.03B
Context32K
Best fitQ8_0 3.6 GB
GGUF
GE
Gemma 3 1b It
Gemma · bartowski
Params1B
Context32K
Best fitBF16 2.7 GB
GGUF
PI
PII Tracer
Pii · perplexity-ai
Params596M
Context32K
Best fitF16 2.6 GB
GGUF
QW
Qwen3.8 3.6 27B blend
Qwen · JetBrains
Params27.3B
Context256K
Best fitQ4_K_M 1.8 GB
ReasoningTool useVision
CH
Chandra Ocr 2
Chandra · prithivMLmods
Params4.84B
Context256K
Best fitQ3_K_S 3.9 GB
ReasoningTool useVision
MI
MiniCPM OCR 1.0
MiniCPM · openbmb
Params752M
Context256K
Best fitF16 2.4 GB
ReasoningTool useVision
QW
Qwen3 Zero Coder Reasoning V2 0.8B NEO EX
Qwen · DavidAU
Params816M
Context40K
Best fitF16 3.5 GB
CodingReasoning
GR
Granite Embedding 107m Multilingual
Granite · bartowski
Params107M
Context512
Best fitF16 0.8 GB
Embedding
QW
Qwen 3.5 27B Derestricted
Qwen · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
GGUF
QW
NasledieLab Qwen3.5 0.8B GPT 6 Astra Luna Fable 5.1 Opus 5.5 distill Karpy
Qwen · aemmeath
Params752M
Context256K
Best fitQ4_K_M 1.4 GB
GGUF
GR
Granite-4.1-3B
Granite · weareapexcreators
Params3.4B
Context128K
Best fitQ5_K_M 3.6 GB
GGUF
QW
Qwen3.5 27B WebNovel Writer Zh
Qwen · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
GGUF
MU
Muse-Glimmer-30B
Muse · AtomicChat
Params27.9B
Context128K
Q4_K_M16.9 GB
ReasoningTool useVision
QW
Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Heretic
Qwen · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
Reasoning
NO
nomic-embed-text-v1
Nomic · nomic-ai
Params137M
Context2K
Best fitF32 1.3 GB
Embedding
LL
Llama-3.1-8B-Instruct
Llama · unsloth
Params8B
Context128K
Best fitIQ1_M 3.9 GB
GGUF
LF
LFM2 2.6B
Lfm · LiquidAI
Params2.57B
Context125K
Best fitQ4_0 4.0 GB
GGUF
QW
Qwen2.5 VL 7B Instruct
Qwen · bartowski
Params7.62B
Context125K
Best fitQ2_K 4.0 GB
Vision
QW
qwen2-1_5b-instruct
Qwen · Qwen
Params1.54B
Context32K
Q4_K_M1.0 GB
GGUF
SM
SmolLM2 1.7B Instruct
SmolLM · bartowski
Params1.71B
Context8K
Best fitQ8_0 3.9 GB
GGUF
QW
Qwen2.5 7b Ins v3
Qwen · bartowski
Params7.62B
Context32K
Best fitQ2_K 4.0 GB
GGUF
DE
deepseek-ai_deepseek-coder-1.3b-instruct
DeepSeek · TheBloke
Params1.35B
Context16K
Best fitQ8_0 3.5 GB
Coding
MI
MinerU2.5 Pro 2605 1.2B
Mineru · jinzhenj
Params494M
Context8K
Best fitQ8_0 1.1 GB
Vision
QW
speakoflow mini
Qwen · SpeakoFlow
Params773M
Context256K
Best fitBF16 2.5 GB
GGUF
QW
Qwen2.5-Omni-7B
Qwen · unsloth
Params7.62B
Context32K
Best fitQ2_K 4.0 GB
Vision
QW
Qwen2.5 3B Instruct Abliterated
Qwen · mradermacher
Params3.4B
Context32K
Best fitQ6_K 3.6 GB
GGUF

Other memory sizes

Keep browsing