GGUF models by Theros

103 GGUF models quantized by Theros on Hugging Face. Sizes run from 40M to 14.8B parameters. The longest trained context is 1000K tokens.

Most downloadedNewestLargestLongest contextSmallest
QW
Qwen2.5 ColdBrew Apex
Qwen · Theros
Params8.78B
Context128K
Q4_K_M5.4 GB
GGUF
CO
Q2.5 ColdBrew Forge Orion Test0
Coldbrew · Theros
Params9.48B
Context128K
Smallest6.8 GB
GGUF
QW
Qwen3 14B VitriousGrey Test0
Qwen · Theros
Params14.8B
Context32K
Q4_K_M9.0 GB
GGUF
PH
Phi 4 ColdBrew 14B Intrepid Test1
Phi · Theros
Params14.7B
Context16K
Q4_K_M9.1 GB
GGUF
CO
Q2.5 ColdBrew 14B Base
Coldbrew · Theros
Params14.8B
Context128K
Q4_K_M9.0 GB
GGUF
CO
Q2.5 ColdBrew Antrax Forge 10B
Coldbrew · Theros
Params9.94B
Context128K
Q4_K_M6.1 GB
GGUF
LL
Llama3.2 ColdBrew 3B Oracle Test0
Llama · Theros
Params3.21B
Context128K
Smallest2.3 GB
GGUF
AR
Arcadia 12B Fusion
Arcadia · Theros
Params12.2B
Context1000K
Q4_K_M7.5 GB
GGUF
AR
Arcadia 12B Fusion Test0
Arcadia · Theros
Params12.2B
Context1000K
Q4_K_M7.5 GB
GGUF
CO
L3 ColdBrew Altair Q4 K M
Coldbrew · Theros
Params8.03B
Context8K
Q4_K_M4.9 GB
GGUF
TE
temp1234 F16
Temp · Theros
Params86M
Context—
Smallest0.2 GB
GGUF
QW
Qwen2.5 ColdBrew R1 Test
Qwen · Theros
Params7.61B
Context128K
Q4_K_M4.7 GB
Reasoning
CO
Q2.5 ColdBrew R1 Atrax Q4 K M
Coldbrew · Theros
Params9.24B
Context128K
Q4_K_M5.7 GB
Reasoning
CH
Q3 ChatThink 9B TestZZ
Chatthink · Theros
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningTool use
CO
L3 ColdBrew Astrid Q4 K M
Coldbrew · Theros
Params8.03B
Context8K
Q4_K_M4.9 GB
GGUF
LL
Llama3.2 ColdBrew 4x3B Argon Test0
Llama · Theros
Params9.55B
Context128K
Smallest6.8 GB
GGUF
SO
SOR ColdBrew 12B Think Base
Sor · Theros
Params12.2B
Context1000K
Q4_K_M7.5 GB
Reasoning
CO
ColdBrew 12B Nemo Orion Test0
Coldbrew · Theros
Params12.2B
Context1000K
Q4_K_M7.5 GB
GGUF
CO
L3 ColdBrew CoT
Coldbrew · Theros
Params8.03B
Context8K
Q4_K_M4.9 GB
Reasoning
CO
Q2.5 ColdBrew R1 Indigo Q4 K M
Coldbrew · Theros
Params9.24B
Context128K
Q4_K_M5.7 GB
Reasoning
CO
ColdBrew Nemo 12B Arcane Fusion Combined Thinker
Coldbrew · Theros
Params12.2B
Context1000K
Q4_K_M7.5 GB
Reasoning
QW
Qwen2.5 ColdBrew R1 test5 Q4 K M
Qwen · Theros
Params7.61B
Context128K
Q4_K_M4.7 GB
Reasoning
CO
L3 ColdBrew SpicyR1 test Q4 K M
Coldbrew · Theros
Params8.03B
Context8K
Q4_K_M4.9 GB
GGUF
CO
Q2.5 ColdBrew R1 Aspera
Coldbrew · Theros
Params9.24B
Context128K
Q4_K_M5.7 GB
ReasoningTool use
QW
Qwen3 ColdBrew 14B Base Test0
Qwen · Theros
Params14.8B
Context40K
Q4_K_M9.0 GB
ReasoningTool use
SP
L3 SpicyOmelettes 10B Test
Spicyomelettes · Theros
Params9.78B
Context8K
Q4_K_M6.0 GB
GGUF
WA
WayChron 12B Test0
Waychron · Theros
Params12.2B
Context1000K
Q4_K_M7.5 GB
GGUF
QW
Qwen3 14B DarkFusion Test0
Qwen · Theros
Params14.8B
Context40K
Q4_K_M9.0 GB
GGUF
LL
Llama3 12B UltraMix Test0
Llama · Theros
Params12.2B
Context128K
Q4_K_M7.4 GB
Tool use
CO
ColdBrew Nemo 12B Arcane Fusion Combined
Coldbrew · Theros
Params12.2B
Context1000K
Q4_K_M7.5 GB
GGUF

About Theros on Hugging Face

Theros has published 103 GGUF models that ggufmodels.com tracks, across families such as Coldbrew, Qwen and Phi. The most downloaded are Qwen2.5 ColdBrew Apex, Q2.5 ColdBrew Forge Orion Test0 and Qwen3 14B VitriousGrey Test0.

The quants Theros publishes most often are Q4_K_M (81), Q5_K_M (17), F16 (4) and Q6_K (2). Their Q4_K_M files range from 4.7 to 9.1 GB. Parameter counts run from 40M to 14.8B, and the longest trained context window is 1000K tokens.

Frequently asked questions

What GGUF models has Theros published?

103 models are listed here, mainly from the Coldbrew, Qwen and Phi families. The most downloaded is Qwen2.5 ColdBrew Apex.

Which quants does Theros publish?

The most common are Q4_K_M (81 models), Q5_K_M (17 models), F16 (4 models) and Q6_K (2 models).

Keep browsing