GGUF models by ZeroXClem

146 GGUF models quantized by ZeroXClem on Hugging Face. Sizes run from 2.03B to 8.95B parameters. The longest trained context is 986K tokens.

Most downloadedNewestLargestLongest contextSmallest
QW
Qwen3.5 9B Fable 5 Quad Stock
Qwen · ZeroXClem
Params8.95B
Context256K
Smallest5.3 GB
ReasoningTool useVision
QW
Qwen3.5 9B Fable 5 Quad Stock
Qwen · ZeroXClem
Params8.95B
Context256K
Smallest9.5 GB
ReasoningTool useVision
QW
Qwen3.5 9B Fable 5 Quad Stock
Qwen · ZeroXClem
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningTool useVision
QW
Qwen3.5 9B Fable 5 Quad Stock
Qwen · ZeroXClem
Params8.95B
Context256K
Smallest4.6 GB
ReasoningTool useVision
QW
Qwen3.5 9B Fable 5 Quad Stock
Qwen · ZeroXClem
Params8.95B
Context256K
Smallest7.4 GB
ReasoningTool useVision
QW
Qwen3.5 9B Fable 5 Quad Stock
Qwen · ZeroXClem
Params8.95B
Context256K
Smallest3.8 GB
ReasoningTool useVision
QW
Qwen3.5 9B Fable 5 Quad Stock
Qwen · ZeroXClem
Params8.95B
Context256K
Smallest6.5 GB
ReasoningTool useVision
QW
Qwen3.5 9B Fable 5 Quad Stock
Qwen · ZeroXClem
Params8.95B
Context256K
Smallest6.3 GB
ReasoningTool useVision
AS
Astral Fusion Neural Happy L3.1 8B Q4 0
Astral · ZeroXClem
Params8.03B
Context8K
Smallest4.7 GB
Roleplay
LL
Llama3.1 TheiaFire DarkFusion 8B Q5 0
Llama · ZeroXClem
Params8.03B
Context128K
Smallest5.6 GB
CodingReasoningRoleplay
ST
Stheno Hercules 3.1 8B Q4 0
Stheno · ZeroXClem
Params8.03B
Context128K
Smallest4.7 GB
GGUF
LL
Llama3.1 BestMix Chem Einstein 8B Q6 K
Llama · ZeroXClem
Params8.03B
Context128K
Smallest6.6 GB
GGUF
QW
Qwen2.5 7B HomerCreative Mix Q5 K M
Qwen · ZeroXClem
Params7.62B
Context32K
Smallest5.4 GB
RoleplayTool use
AS
Astral Fusion Neural Happy L3.1 8B Q8 0
Astral · ZeroXClem
Params8.03B
Context8K
Smallest8.5 GB
Roleplay
LL
L3.1 Purosani 2 8B
Llama · ZeroXClem
Params8.03B
Context128K
Smallest6.6 GB
RoleplayTool use
LL
Llama3.1 TheiaFire DarkFusion 8B Q4 K M
Llama · ZeroXClem
Params8.03B
Context128K
Q4_K_M4.9 GB
CodingReasoningRoleplay
LL
Llama3.1 8B Titanium Forge Q4 0
Llama · ZeroXClem
Params8.03B
Context128K
Smallest4.7 GB
CodingRoleplayTool use
AS
Astral Fusion Neural Happy L3.1 8B Q6 K
Astral · ZeroXClem
Params8.03B
Context8K
Smallest6.6 GB
Roleplay
LL
L3.1 Purosani 2 8B
Llama · ZeroXClem
Params8.03B
Context128K
Q4_K_M4.9 GB
RoleplayTool use
QW
Qwen2.5 7B
Qwen · ZeroXClem
Params7.62B
Context128K
Smallest5.3 GB
Tool use
LL
Llama3.1 8B Titanium Forge Q4 K M
Llama · ZeroXClem
Params8.03B
Context128K
Q4_K_M4.9 GB
CodingRoleplayTool use
LL
Llama3.1 TheiaFire DarkFusion 8B Q4 K S
Llama · ZeroXClem
Params8.03B
Context128K
Smallest4.7 GB
CodingReasoningRoleplay
QW
Qwen2.5 7B HomerAnvita NerdMix Q4 0
Qwen · ZeroXClem
Params7.62B
Context32K
Smallest4.4 GB
RoleplayTool use
QW
Qwen3 4B Sky High Hermes
Qwen · ZeroXClem
Params4.02B
Context256K
Smallest3.3 GB
ReasoningTool use
QW
Qwen3 4B Wrist On Hermes
Qwen · ZeroXClem
Params4.41B
Context256K
Smallest4.7 GB
ReasoningTool use
LL
Llama3.1 TheiaFire DarkFusion 8B Q5 K M
Llama · ZeroXClem
Params8.03B
Context128K
Smallest5.7 GB
CodingReasoningRoleplay
AS
Astral Fusion Neural Happy L3.1 8B Q5 0
Astral · ZeroXClem
Params8.03B
Context8K
Smallest5.6 GB
Roleplay
QW
Qwen2.5 7B HomerCreative Mix Q5 0
Qwen · ZeroXClem
Params7.62B
Context32K
Smallest5.3 GB
RoleplayTool use
QW
Qwen2.5 7B
Qwen · ZeroXClem
Params7.62B
Context128K
Q4_K_M4.7 GB
Tool use
QW
Qwen2.5 7B HomerCreative Mix Q8 0
Qwen · ZeroXClem
Params7.62B
Context32K
Smallest8.1 GB
RoleplayTool use

About ZeroXClem on Hugging Face

ZeroXClem has published 146 GGUF models that ggufmodels.com tracks, across families such as Llama, Qwen and Astral. The most downloaded are Qwen3.5 9B Fable 5 Quad Stock, Qwen3.5 9B Fable 5 Quad Stock and Qwen3.5 9B Fable 5 Quad Stock.

The quants ZeroXClem publishes most often are Q4_K_M (36), Q6_K (24), Q8_0 (21) and Q5_0 (18). Their Q4_K_M files range from 1.3 to 5.6 GB. Parameter counts run from 2.03B to 8.95B, and the longest trained context window is 986K tokens.

Frequently asked questions

What GGUF models has ZeroXClem published?

146 models are listed here, mainly from the Llama, Qwen and Astral families. The most downloaded is Qwen3.5 9B Fable 5 Quad Stock.

Which quants does ZeroXClem publish?

The most common are Q4_K_M (36 models), Q6_K (24 models), Q8_0 (21 models) and Q5_0 (18 models).

Keep browsing