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.
QU
QuasiStarSynth 12B Absolute Heresy
Quasistarsynth · mradermacher
Params12.2B
Context1000K
Q4_K_M7.5 GB
AI
airoboros-dpo-70b-3.3
Airoboros · mradermacher
Params70.6B
Context8K
Q4_K_M42.5 GB
MU
multimaster-7b-v2
Multimaster · mradermacher
Params35.4B
Context32K
Q4_K_M21.4 GB
MI
Mistral-Small-3.2-24B-Instruct-2506
Mistral · Sci-fi-vy
Params23.6B
Context128K
Q4_K_M14.3 GB
GE
Gemma 3 12b It Vl GLM 4.7 Flash INSTRUCT Thinking Hybrid Heretic Uncensored
Gemma · mradermacher
Params12.8B
Context128K
Q4_K_M7.9 GB
CO
CogniDet
Cognidet · mradermacher
Params8.03B
Context128K
Q4_K_M4.9 GB
DE
WR30a Deep 7B 0711
Deep · mradermacher
Params7.62B
Context125K
Q4_K_M4.7 GB
MI
Mistral 7B Instruct v0.2
Mistral · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
MI
Minecraftism 4B Thinking Uncensored
Minecraftism · mradermacher
Params4.02B
Context256K
Q4_K_M2.5 GB
LL
Llama 3.1 EIRAI 8B Prob i1
Llama · mradermacher
Params8B
Context8K
Q4_K_M4.9 GB
SM
Smart Commit 0.5b v0.3
Smart · mradermacher
Params494M
Context32K
Q4_K_M0.4 GB
ZE
BA Zephyria 39b i1
Zephyria · mradermacher
Params39B
Context128K
Q4_K_M23.4 GB
LL
OpenBioLLM Llama3 70B i1
Llama · mradermacher
Params70.6B
Context8K
Q4_K_M42.5 GB
PA
MS3.2 PaintedFantasy Visage v3 34B Ultra Uncensored Heretic
Paintedfantasy · mradermacher
Params34.1B
Context128K
Q4_K_M20.7 GB
JA
Jais Adapted 70b Chat
Jais · mradermacher
Params69.5B
Context4K
Q4_K_M41.8 GB
NE
NeuralMarcoro14 7B i1
Neuralmarcoro · mradermacher
Params7.24B
Context32K
Q4_K_M4.4 GB
QW
Qwen3.8 27B Heretic v2
Qwen · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
GE
Gemma 4 26B A4B It
Gemma · mradermacher
Params25.2B
Context256K
Q4_K_M16.8 GB
GE
Gemma 4 19B A4B It The DECKARD Thinking
Gemma · mradermacher
Params18.4B
Context256K
Q4_K_M12.3 GB
IN
InternVL3_5 8B
Internvl · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
LL
LLaMA3 iterative DPO final
Llama · bartowski
Params8.03B
Context8K
Q4_K_M4.9 GB
QW
Qwen3 VL Reranker 8B
Qwen · mradermacher
Params8.19B
Context256K
Q4_K_M5.0 GB
QW
Bokap Betawi v0 Qwen2.5 7B fp16 i1
Qwen · mradermacher
Params7.62B
Context128K
Q4_K_M4.7 GB
PH
AS Pharade 8B
Pharade · mradermacher
Params8B
Context32K
Q4_K_M4.9 GB
GP
Gpt Oss 20b Abliterated
gpt-oss · mradermacher
Params20.9B
Context128K
Q4_K_M15.8 GB
LL
calme 2.4 llama3 70b i1
Llama · mradermacher
Params70.6B
Context8K
Q4_K_M42.5 GB
XI
XiaoHong v1
Xiaohong · mradermacher
Params8.2B
Context40K
Q4_K_M5.0 GB
VI
Vims2 7B
Vims · mradermacher
Params7.62B
Context128K
Q4_K_M4.7 GB
DE
Delirium v1 i1
Delirium · mradermacher
Params9.24B
Context8K
Q4_K_M5.8 GB
QW
X Coder RL Qwen2.5 7B
Qwen · mradermacher
Params7.62B
Context32K
Q4_K_M4.7 GB