Tiny GGUF models
62726 GGUF models with 3B parameters or fewer. Sizes run from 1M to 3.5B parameters. The longest trained context is 795898K tokens.
ST
Strands
Strands · fabricant451
Params1.88B
Context4K
Smallest2.0 GB
QW
Qwen2 1.5B
Qwen · artificialguybr
Params1.78B
Context128K
Q4_K_M1.1 GB
EX
EXAONE 3.5 2.4B Instruct
Exaone · bartowski
Params2.67B
Context32K
Q4_K_M1.6 GB
QW
Qwen3 VL 2B Thinking Abliterated v1
Qwen · prithivMLmods
Params1.72B
Context256K
Q4_K_M1.1 GB
GE
Gemma 2 2b Jpn It
Gemma · mradermacher
Params2.61B
Context8K
Q4_K_M1.7 GB
BI
bitnet embedding 270m
Bitnet · microsoft
Params268M
Context32K
Smallest0.4 GB
BA
Baguettotron
Baguettotron · PleIAs
Params321M
Context4K
Q4_K_M0.2 GB
OW
OWoTGPT 1.3
Owotgpt · Pomni
Params124M
Context1K
Q4_K_M0.1 GB
QW
Qwen2 VL 2B Abliterated Caption It
Qwen · mradermacher
Params1.54B
Context32K
Q4_K_M1.0 GB
SH
Shruti Soft 2.6b
Shruti · mradermacher
Params2.7B
Context128K
Q4_K_M1.7 GB
QW
Qwen3.5 2B
Qwen · mradermacher
Params1.88B
Context256K
Q4_K_M1.3 GB
NU
Nuqta Arabic OCR 0.8B
Nuqta · SalahALHaismawi
Params810M
Context256K
Q4_K_M0.6 GB
QW
Qwen3.5 2B Ideal TSUNDERE Loli Girl Japanese v1
Qwen · mradermacher
Params1.88B
Context256K
Q4_K_M1.3 GB
GE
Gemma 2 2b Chinese it
Gemma · QuantFactory
Params2.61B
Context8K
Q4_K_M1.7 GB
GR
granite vision 3.2 2b
Granite · ibm-research
Params2.53B
Context16K
Q4_K_M1.5 GB
MI
Mina 1B Frog RL Zero
Mina · Trivnpk
Params1.24B
Context128K
Q4_K_M0.8 GB
LI
lingbot map
Lingbot · Asher-1
Params1.16B
Context—
Smallest0.0 GB
MO
MOSS TTS Realtime
Moss · Yanlis
Params2.08B
Context40K
Q4_K_M1.1 GB
QW
Qwen3.5 2B multiharness RL
Qwen · igorktech
Params1.88B
Context256K
Q4_K_M1.3 GB
MI
Ministral 3B Heresy Unfiltered
Mistral · mradermacher
Params3.43B
Context256K
Q4_K_M2.1 GB
NO
Nomic Embed Text v1.5
Nomic · gaianet
Params137M
Context2K
Q4_K_M0.1 GB
QW
Qwen2.5 0.5B
Qwen · mradermacher
Params494M
Context32K
Q4_K_M0.4 GB
NE
Raj 0
Nemotron · aungzaythant
Params637M
Context—
Smallest0.7 GB
MI
Ministral 3 3B Claude Opus Reasoning Distilled
Mistral · mradermacher
Params3.43B
Context256K
Q4_K_M2.1 GB
QW
Qwen2.5 1.5B VibeThinker Heretic Uncensored Abliterated
Qwen · mradermacher
Params1.54B
Context128K
Q4_K_M1.0 GB
QW
ThinCtx Qwen3.5 2B
Qwen · mradermacher
Params1.88B
Context256K
Q4_K_M1.3 GB
ED
Edceleste Decider 0.8b
Edceleste · mradermacher
Params752M
Context256K
Q4_K_M0.5 GB
QW
Qwen2.5 VL 3B Instruct Heretic
Qwen · mradermacher
Params3.09B
Context125K
Q4_K_M1.9 GB
GR
Ibm Granite_Granite 4.1 3b
Granite · lmstudio-community
Params3.4B
Context128K
Q4_K_M2.1 GB
IM
ImageShield MMCF 0.8B
Imageshield · prithivMLmods
Params752M
Context256K
Q4_K_M0.5 GB