IQ2_XXS GGUF models
Importance-matrix 2-bit quant. Extreme compression with a large quality cost. 26647 models have a IQ2_XXS file. Sizes run from 2M to 2780B parameters. The longest trained context is 10240K tokens.
KI
Kimi-K2-Instruct
Kimi · unsloth
Params1030B
Context128K
Q4_K_M620.8 GB
QW
Qwen2.5-Omni-7B
Qwen · unsloth
Params7.62B
Context32K
Q4_K_M4.7 GB
GE
Gemma 4 26B A4B StyleTune V2
Gemma · mradermacher
Params26B
Context256K
Q4_K_M17.2 GB
MI
Ministral-3-14B-Instruct-2512
Mistral · unsloth
Params13.5B
Context256K
Q4_K_M8.2 GB
QW
Qwen3.5-35B-A3B
Qwen · unsloth
Params35.5B
Context256K
Q4_K_M22.6 GB
ME
Medgemma-1.5-4B-It
Medgemma · unsloth
Params3.88B
Context128K
Q4_K_M2.5 GB
HO
K2 Horizon MoVA 36B A4B
Horizon · NANI-Nithin
Params37.4B
Context512K
Q4_K_M22.4 GB
DE
Huihui DeepSeek V4.1 Flash 1017 abliterated
DeepSeek · huihui-ai
Params1.68B
Context1024K
Smallest0.6 GB
DE
Deepseek-R1-0528
DeepSeek · unsloth
Params670B
Context160K
Q4_K_M404.9 GB
GL
Huihui GLM 4.7 Flash Abliterated
GLM · mradermacher
Params30B
Context198K
Q4_K_M18.1 GB
LA
Laguna-S-2.1
Laguna · unsloth
Params118B
Context256K
Q4_K_M73.1 GB
MI
minmax
Minmax · Kutches
Params20.1B
Context—
Q4_K_M11.5 GB
QW
Suri Qwen 3.8 27B Companion
Qwen · SpaceTimee
Params27.3B
Context256K
Q4_K_M16.5 GB
MI
MiniMax M3.1 Flash Maya
Minimax · peasantsmith
Params426B
Context1024K
Smallest93.0 GB
SM
Smollm3-3B
SmolLM · unsloth
Params3.08B
Context64K
Q4_K_M1.9 GB
NE
Nvidia-Nemotron-3-Nano-Omni-30B-A3B-Reasoning
Nemotron · unsloth
Params31.6B
Context1024K
Q4_K_M23.9 GB
MI
MiniMax-M3.1-Flash
Minimax · LyJonathan
Params426B
Context1024K
Smallest94.8 GB
QW
Qwen3 VL 32B Instruct
Qwen · bartowski
Params32.8B
Context256K
Q4_K_M19.8 GB
MI
Ministral-3-8B-Reasoning-2512
Mistral · unsloth
Params8.49B
Context256K
Q4_K_M5.2 GB
MI
MiniMax M3.1 Flash
Minimax · unsloth
Params426B
Context1024K
Smallest0.5 GB
QW
Qwen3-4B-Thinking-2507
Qwen · unsloth
Params4B
Context256K
Q4_K_M2.5 GB
GO
Goetia 26B A4B v1.7
Goetia · mradermacher
Params26B
Context256K
Q4_K_M17.2 GB
LF
LFM2.5-24B-A2B
Lfm · NANI-Nithin
Params24B
Context128K
Q4_K_M14.5 GB
GE
Gemma 4 Orthia 31B
Gemma · mradermacher
Params32.1B
Context256K
Q4_K_M19.5 GB
DE
DeepSeek-V2-Lite
DeepSeek · legraphista
Params15.7B
Context160K
Smallest5.0 GB
DE
DeepSeek R1 Distill Llama 70B
DeepSeek · bartowski
Params70.6B
Context128K
Q4_K_M42.5 GB
GE
Gemma 4 31B It Heretic
Gemma · mradermacher
Params30.7B
Context256K
Q4_K_M18.7 GB
QW
Qwen2.5 VL 32B Instruct Unredacted MAX
Qwen · mradermacher
Params32.8B
Context125K
Q4_K_M19.9 GB
QW
Qwen3.8 27B Uncensored Heretic Abliterated
Qwen · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
LL
Meta llama_Meta Llama 3 70B Instruct
Llama · ThomasBaruzier
Params70.6B
Context8K
Q4_K_M42.5 GB