GGUF models by aixk

42 GGUF models quantized by aixk on Hugging Face. Sizes run from 50M to 778M parameters. The longest trained context is 1024K tokens.

Most downloadedNewestLargestLongest contextSmallest
QW
Qwen2.5 Coder 0.5B Instruct
Qwen · aixk
Params494M
Context32K
Q4_K_M0.4 GB
CodingTool use
HA
haru 390m
Haru · aixk
Params391M
Context8K
Q4_K_M0.3 GB
GGUF
QW
Qwen2.5 0.5B Instruct
Qwen · aixk
Params494M
Context32K
Q4_K_M0.4 GB
Tool use
AU
Aurora 0.5B Volcano_final
Aurora · aixk
Params670M
Context2K
Q4_K_M0.4 GB
Coding
LF
LFM2.5 230M MINI 2bit 3bit 4bit
Lfm · aixk
Params179M
Context125K
Q4_K_M0.1 GB
ReasoningTool use
HA
haru 180m
Haru · aixk
Params184M
Context8K
Q4_K_M0.1 GB
GGUF
KO
Kori 390M
Kori · aixk
Params391M
Context8K
Smallest0.8 GB
GGUF
QW
Qwen2.5 0.5B Korean OpenThoughts
Qwen · aixk
Params494M
Context32K
Q4_K_M0.4 GB
Tool use
LF
LFM2.5 230M TINY
Lfm · aixk
Params196M
Context125K
Q4_K_M0.1 GB
ReasoningTool use
LF
LFM2.5 230M 1bit micro
Lfm · aixk
Params179M
Context125K
Smallest0.0 GB
ReasoningTool use
CU
CustomConversationalAI
Customconversationalai · aixk
Params275M
Context256
Q4_K_M0.2 GB
GGUF
SM
SmolLM2 360M Instruct
SmolLM · aixk
Params362M
Context8K
Q4_K_M0.3 GB
GGUF
BI
0.7B Bilingual ChatML
Bilingual · aixk
Params778M
Context2K
Q4_K_M0.5 GB
GGUF
SM
SmolLM2 135M
SmolLM · aixk
Params135M
Context8K
Q4_K_M0.1 GB
GGUF
LF
LFM2.5 230M 16k MINI
Lfm · aixk
Params179M
Context125K
Smallest0.1 GB
ReasoningTool use
GR
Granite 4.0 H 350m Base
Granite · aixk
Params340M
Context1024K
Q4_K_M0.2 GB
GGUF
NO
NovaCore Final
Novacore · aixk
Params—
Context—
Q4_K_M0.0 GB
GGUF
CU
custom model
Custom · aixk
Params50M
Context512
Q4_K_M0.0 GB
GGUF
KO
kosmo TEST
Kosmo · aixk
Params167M
Context2K
Smallest0.3 GB
GGUF
NE
Nexus 0.7b Final
Nexus · aixk
Params778M
Context2K
Q4_K_M0.5 GB
GGUF
CO
CORE X 0.7B Dialogue v1.2
Core · aixk
Params348M
Context2K
Q4_K_M0.2 GB
GGUF
CO
X Core Chat V1
Core · aixk
Params—
Context—
Smallest0.1 GB
GGUF
SS
ssai stack 0 10B
Ssai · aixk
Params118M
Context128
Smallest0.2 GB
GGUF
LF
LFM2.5 230M 2bit 72.6MB
Lfm · aixk
Params179M
Context125K
Smallest0.1 GB
ReasoningTool use
KO
kosine GGUF F16
Kosine · aixk
Params76M
Context512
Smallest0.2 GB
GGUF
SP
spirit 0 2B
Spirit · aixk
Params304M
Context128
Q4_K_M0.2 GB
GGUF
CU
CustomKoChatbot
Customkochatbot · aixk
Params58M
Context512
Q4_K_M0.0 GB
GGUF
KO
korean test
Korean · aixk
Params65M
Context512
Smallest0.1 GB
GGUF
TW
twiny stack L03
Twiny · aixk
Params75M
Context128
Smallest0.2 GB
GGUF
NO
Nova 0.6B Turbo Companion final
Nova · aixk
Params495M
Context2K
Smallest1.0 GB
GGUF

About aixk on Hugging Face

aixk has published 42 GGUF models that ggufmodels.com tracks, across families such as Lfm, Core and Twiny. The most downloaded are Qwen2.5 Coder 0.5B Instruct, haru 390m and Qwen2.5 0.5B Instruct.

The quants aixk publishes most often are Q4_K_M (27), F16 (21), Q3_K_M (7) and Q2_K (5). Their Q4_K_M files range from 0.0 to 1.3 GB. Parameter counts run from 50M to 778M, and the longest trained context window is 1024K tokens.

Frequently asked questions

What GGUF models has aixk published?

42 models are listed here, mainly from the Lfm, Core and Twiny families. The most downloaded is Qwen2.5 Coder 0.5B Instruct.

Which quants does aixk publish?

The most common are Q4_K_M (27 models), F16 (21 models), Q3_K_M (7 models) and Q2_K (5 models).

Keep browsing