GGUF models by calcuis
50 GGUF models quantized by calcuis on Hugging Face. Sizes run from 123M to 20.4B parameters. The longest trained context is 128K tokens.
WA
wan 1.3b
Wan · calcuis
Params127M
Context—
Q4_K_M1.0 GB
ST
sd3.5 large
Stable · calcuis
Params8.15B
Context—
Smallest4.8 GB
LT
ltxv0.9.6
Ltx · calcuis
Params1.92B
Context—
Q4_K_M1.4 GB
AC
ace
Ace · calcuis
Params3.31B
Context—
Q4_K_M0.1 GB
QW
qwen image edit plus
Qwen · calcuis
Params14.3B
Context—
Q4_K_M11.7 GB
CH
chatterbox
Chatterbox · calcuis
Params264M
Context—
Q4_K_M0.3 GB
FL
krea
Flux · calcuis
Params11.9B
Context—
Q4_K_M7.1 GB
FL
flux1
Flux · calcuis
Params11.9B
Context—
Q4_K_M7.0 GB
ST
sd3.5 medium
Stable · calcuis
Params2.47B
Context—
Q4_K_M1.8 GB
QW
qwen image edit
Qwen · calcuis
Params20.4B
Context—
Q4_K_M11.7 GB
LU
lumina
Lumina · calcuis
Params2.61B
Context—
Q4_K_M1.5 GB
QW
qwen image
Qwen · calcuis
Params20.4B
Context—
Q4_K_M11.7 GB
WA
wan s2v
Wan · calcuis
Params16.3B
Context—
Q4_K_M13.9 GB
HU
hyvid
Hunyuanvideo · calcuis
Params12.8B
Context—
Q4_K_M7.9 GB
HU
hy3d
Hunyuan · calcuis
Params1.11B
Context—
Q4_K_M0.6 GB
PI
pixart
Pixart · calcuis
Params611M
Context—
Q4_K_M1.0 GB
BA
bagel
Bagel · calcuis
Params14.6B
Context—
Q4_K_M8.9 GB
DE
DeepSeek R1 Distill Llama 8B
DeepSeek · calcuis
Params8B
Context128K
Q4_K_M4.9 GB
PH
phi4
Phi · calcuis
Params14.7B
Context16K
Q4_K_M8.9 GB
LI
sd3.5 lite
Lite · calcuis
Params2.47B
Context—
Q4_K_M1.8 GB
HU
hyvid i2v
Hunyuanvideo · calcuis
Params123M
Context—
Q4_K_M7.7 GB
SK
sketch
Sketch · calcuis
Params11.9B
Context—
Smallest4.0 GB
HU
hunyuan
Hunyuanvideo · calcuis
Params12.8B
Context—
Q4_K_M7.9 GB
LT
ltxv0.9.7
Ltx · calcuis
Params13B
Context—
Q4_K_M9.1 GB
OM
omnigen2
Omnigen · calcuis
Params3.97B
Context—
Q4_K_M6.0 GB
ST
sd3.5 large controlnet
Stable · calcuis
Params8.15B
Context—
Smallest4.8 GB
HI
higgs
Higgs · calcuis
Params5.77B
Context—
Q4_K_M4.5 GB
HU
humo
Humo · calcuis
Params1.74B
Context—
Q4_K_M1.2 GB
LT
ltxv0.9.5
Ltx · calcuis
Params1.92B
Context—
Q4_K_M1.1 GB
CO
cow encoder
Cow · calcuis
Params2.62B
Context—
Smallest0.4 GB
About calcuis on Hugging Face
calcuis has published 50 GGUF models that ggufmodels.com tracks, across families such as Stable, Qwen and Phi. The most downloaded are wan 1.3b, sd3.5 large and ltxv0.9.6.
The quants calcuis publishes most often are Q8_0 (41), Q4_0 (41), Q5_0 (39) and Q6_K (38). Their Q4_K_M files range from 0.1 to 13.9 GB. Parameter counts run from 123M to 20.4B, and the longest trained context window is 128K tokens.
Frequently asked questions
What GGUF models has calcuis published?
50 models are listed here, mainly from the Stable, Qwen and Phi families. The most downloaded is wan 1.3b.
Which quants does calcuis publish?
The most common are Q8_0 (41 models), Q4_0 (41 models), Q5_0 (39 models) and Q6_K (38 models).