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
IM
Imagine v10
Imagine · Interchained
Params1.35B
Context16K
Q4_K_M0.9 GB
Coding
BA
Baked Egglet 2B Model Stock
Baked · matrixportalx
Params2.61B
Context8K
Q4_K_M1.7 GB
GGUF
QW
ONIVO speech correction v2
Qwen · minomonchan
Params752M
Context256K
Q4_K_M0.5 GB
Reasoning
LF
LFM3 1.7B
Lfm · prithivMLmods
Params1.73B
Context256K
Q4_K_M1.1 GB
CodingReasoningTool use
GL
GLM 5.3 0.6B A0.4B
GLM · mradermacher
Params622M
Context1024K
Q4_K_M0.4 GB
ReasoningTool use
TI
TinyLlama 1.1B 32k Instruct
Tinyllama · mradermacher
Params1.1B
Context32K
Q4_K_M0.7 GB
GGUF
QW
Qwen3.5-2B
Qwen · Genrius-AI
Params988M
Context256K
Smallest0.8 GB
ReasoningTool use
LL
Dolphin3.0 Llama3.1 1B abliterated
Llama · NexesQuants
Params1.24B
Context128K
Smallest0.9 GB
GGUF
VA
Vakeel 8B v2 mini
Vakeel · mradermacher
Params1.83B
Context8K
Q4_K_M1.1 GB
GGUF
MI
minimax h3
Minimax · convertor
Params151M
Context—
Smallest0.3 GB
GGUF
LL
Llama 3.2 3B Instruct
Llama · SanctumAI
Params3.21B
Context128K
Q4_K_M2.0 GB
Tool use
LF
LFM2.5-Audio-1.5B
Lfm · Mungert
Params1.17B
Context125K
Q4_K_M0.7 GB
Tool use
CO
C4AI Dpo Model
Command R · mradermacher
Params135M
Context2K
Q4_K_M0.1 GB
GGUF
EX
EXAONE 4.0 1.2B
Exaone · bartowski
Params1.28B
Context64K
Q4_K_M0.8 GB
ReasoningTool use
LL
Hermes 3 Llama 3.2 3b Base Fft Chatml Rerun 8xh100 E4 2024 11 24 4k Steps
Llama · mradermacher
Params3.21B
Context128K
Q4_K_M2.0 GB
RoleplayTool use
GL
GLM 5.3 Flash UNCENSORED MTP Maya L
GLM · Adolphsson
Params564M
Context1024K
Smallest156.3 GB
Vision
LL
Llama 3.2 3B Instruct Roleplay Tuned
Llama · mradermacher
Params3.21B
Context128K
Q4_K_M2.0 GB
RoleplayTool use
QW
OpenHermes Qwen1.5 1.8B i1
Qwen · mradermacher
Params1.84B
Context32K
Q4_K_M1.2 GB
Tool use
FA
Falcon H1 Tiny Multilingual 100M Instruct
Falcon · mradermacher
Params108M
Context256K
Q4_K_M0.1 GB
Tool use
QW
Qwen3 Shining Lucy CODER 2.4B E32
Qwen · mradermacher
Params2.43B
Context40K
Q4_K_M1.5 GB
CodingReasoningTool use
VE
Venusaur i1
Venusaur · mradermacher
Params16M
Context512
Q4_K_M0.0 GB
Embedding
LL
Llama 3.2 3B Instruct
Llama · MCZK
Params3.21B
Context128K
Q4_K_M2.0 GB
Tool use
ST
StartLux Decision v1 0.8B
Startlux · mradermacher
Params773M
Context256K
Q4_K_M0.5 GB
ReasoningTool useVision
MO
Moonshine Base (Korean)
Moonshine · handy-computer
Params62M
Context—
Smallest0.1 GB
GGUF
QW
Qwen2.5-1.5B-Instruct
Qwen · Mungert
Params1.54B
Context32K
Q4_K_M1.0 GB
Tool use
LL
Llama 3.2 1b Abliterated
Llama · mradermacher
Params1.24B
Context128K
Q4_K_M0.8 GB
Tool use
LL
Llama 3B Coder
Llama · koshuro
Params3.21B
Context128K
Q4_K_M2.0 GB
CodingTool use
QW
Qwen3.5-4B
Qwen · Genrius-AI
Params1.7B
Context256K
Smallest1.5 GB
ReasoningTool use
LO
Loopl 0.8b
Loopl · mradermacher
Params752M
Context256K
Q4_K_M0.5 GB
ReasoningTool useVision
NO
NovelAI-LM-3B-508k
Novelai · NovelAI
Params3.04B
Context2K
Q4_K_M2.0 GB
GGUF

Other sizes

Keep browsing