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.

Most downloadedNewestLargestLongest contextSmallest
LF
LFM3 1.7B Uncensored LocalAI
Lfm · weareapexcreators
Params1.73B
Context256K
Q4_K_M1.1 GB
ReasoningTool use
IG
Igea-1B-instruct
Igea · mradermacher
Params1.01B
Context16K
Q4_K_M0.6 GB
GGUF
QW
Qwen Qwen3 0.6B
Qwen · tensorblock
Params752M
Context40K
Smallest0.3 GB
ReasoningTool use
LL
llama 3.2 3b gsm8k
Llama · Anirudhpatil367
Params3.21B
Context128K
Q4_K_M2.0 GB
Tool use
QW
Qwen1.5-1.8B-Chat
Qwen · second-state
Params1.84B
Context32K
Q4_K_M1.2 GB
GGUF
DE
demodokos foundry tts v4
Demodokos · cmp-nct
Params909M
Context—
Smallest0.0 GB
GGUF
EX
EXAONE 3.5 2.4B Instruct
Exaone · LGAI-EXAONE
Params2.67B
Context32K
Q4_K_M1.6 GB
GGUF
AN
anima aesthetic v1.1
Anima · E-stick
Params2.09B
Context—
Smallest2.2 GB
GGUF
GA
GameScaling 0.8B S3
Gamescaling · mradermacher
Params1.01B
Context256K
Q4_K_M0.7 GB
ReasoningTool useVision
BO
Boris 1.3 75M Instruct
Boris · opencerebral
Params77M
Context1K
Q4_K_M0.1 GB
GGUF
DA
H2O Danube3 500m Base
Danube · mradermacher
Params514M
Context8K
Q4_K_M0.3 GB
GGUF
RE
reazonspeech nemo v2
Reazonspeech · cstr
Params619M
Context—
Smallest0.4 GB
GGUF
LF
LFM2.5 2.6B fable5 coding agent
Lfm · AyoubChLin
Params2.7B
Context128K
Q4_K_M1.7 GB
CodingReasoningTool use
PO
polka 1.1b chat
Polka · mradermacher
Params1.15B
Context4K
Q4_K_M0.7 GB
GGUF
QW
ruadapt qwen2.5 3B ext u48 instruct v4
Qwen · RefalMachine
Params3.08B
Context32K
Q4_K_M1.9 GB
Tool use
QW
Qwen_Qwen3.5 0.8B
Qwen · notschmee
Params752M
Context256K
Q4_K_M0.6 GB
ReasoningTool useVision
QW
Qwen3.5-2B
Qwen · TheStageAI
Params1.88B
Context256K
Q4_K_M1.1 GB
ReasoningTool use
QW
Qwen3.5 2B Ideal TSUNDERE Loli Girl Japanese v1
Qwen · mradermacher
Params1.88B
Context256K
Q4_K_M1.3 GB
ReasoningTool useVision
QW
Qwen3 Zero Coder Reasoning Brainstorm5x 0.9B
Qwen · mradermacher
Params879M
Context40K
Q4_K_M0.6 GB
CodingReasoningTool use
QW
jinaai/test-qwen25-coder-jina-code-embeddings-0.5b
Qwen · jinaai
Params494M
Context32K
Smallest0.3 GB
CodingEmbedding
QW
Model Qwen Qwen2.5 1.5B Instruct Cache
Qwen · Alcoft
Params1.54B
Context32K
Q4_K_M1.0 GB
Tool use
QW
Qwen2.5 1.5B Instruct
Qwen · duyntnet
Params1.54B
Context32K
Q4_K_M1.0 GB
GGUF
LL
Llama Guard 3 1B
Llama · QuantFactory
Params1.5B
Context128K
Q4_K_M1.0 GB
GGUF
DY
DynaGuard AIO
Dynaguard · prithivMLmods
Params1.72B
Context40K
Q4_K_M1.1 GB
ReasoningTool use
MI
MiniCPM5 1B
MiniCPM · eaddario
Params1.08B
Context128K
Smallest0.0 GB
GGUF
PR
prem 1B SQL
Prem · QuantFactory
Params1.35B
Context16K
Q4_K_M0.9 GB
GGUF
MI
Super Slop Machina RP XXL Small 3B 2608
Mistral · Indexnusrefather
Params3.43B
Context256K
Q4_K_M2.1 GB
RoleplayTool useVision
QW
Qwen2.5 Coder 0.5B Instruct
Qwen · second-state
Params494M
Context32K
Q4_K_M0.4 GB
CodingTool use
HA
Hare-1.1B-base
Hare · mradermacher
Params1.12B
Context32K
Q4_K_M0.7 GB
GGUF
SP
speecht5 tts
Speecht · cstr
Params157M
Context—
Smallest0.3 GB
GGUF

Other sizes

Keep browsing