GGUF models by WhiskyAKM

35 GGUF models quantized by WhiskyAKM on Hugging Face. Sizes run from 1.01B to 177B parameters. The longest trained context is 1024K tokens.

Most downloadedNewestLargestLongest contextSmallest
LF
LFM2.5 24B A2B
Lfm · WhiskyAKM
Params24B
Context128K
Smallest13.6 GB
GGUF
GE
Gemma 4 E2B it qat NVFP4
Gemma · WhiskyAKM
Params4.63B
Context128K
Smallest0.1 GB
Vision
GE
Gemma 4 26B A4B it qat NVFP4
Gemma · WhiskyAKM
Params25.2B
Context256K
Smallest0.3 GB
Vision
LI
Ling 3.0 Tiny
Ling · WhiskyAKM
Params7.89B
Context128K
Q4_K_M4.8 GB
ReasoningTool use
QW
Qwen3.8 Flash Next
Qwen · WhiskyAKM
Params177B
Context256K
Smallest2.4 GB
ReasoningTool useVision
NE
Nemotron 3.5 Bolt E4B NVFP4
Nemotron · WhiskyAKM
Params7.52B
Context64K
Smallest4.3 GB
ReasoningTool useVision
MI
MiniCPM V 4.7 35B A3B
MiniCPM · WhiskyAKM
Params34.7B
Context256K
Smallest19.7 GB
ReasoningTool useVision
GE
Gemma 4 E2B It Qat Q4_0 Unquantized
Gemma · WhiskyAKM
Params4.63B
Context128K
Q4_K_M3.4 GB
ReasoningTool useVision
MI
MiniCPM5 2B
MiniCPM · WhiskyAKM
Params2.52B
Context128K
Q4_K_M0.2 GB
ReasoningTool use
LI
Ling 3.0 Flash
Ling · WhiskyAKM
Params127B
Context256K
Q4_K_M77.0 GB
ReasoningTool use
G9
G9v3 3B
WhiskyAKM
Params2.99B
Context128K
Q4_K_M1.8 GB
ReasoningTool use
GE
Gemma 4 12B it qat NVFP4
Gemma · WhiskyAKM
Params11.9B
Context256K
Smallest0.3 GB
Vision
QW
Qwen3.8 27B NVFP4
Qwen · WhiskyAKM
Params26.9B
Context256K
Smallest2.0 GB
ReasoningTool useVision
QW
Qwen3.6 35B A3B
Qwen · WhiskyAKM
Params35.5B
Context256K
Q4_K_M21.7 GB
ReasoningTool useVision
SP
Spark X2.6 4B
Spark · WhiskyAKM
Params4.11B
Context1024K
Smallest2.3 GB
ReasoningTool use
GE
Gemma 4 12B It Qat Q4_0 Unquantized
Gemma · WhiskyAKM
Params11.9B
Context256K
Q4_K_M7.4 GB
ReasoningTool useVision
GE
Gemma 4 E4B it qat NVFP4
Gemma · WhiskyAKM
Params7.46B
Context128K
Smallest0.1 GB
Vision
NE
NVIDIA Nemotron 3.5 Lightning 30B A3B
Nemotron · WhiskyAKM
Params32.9B
Context1024K
Q4_K_M25.4 GB
ReasoningTool use
MI
MiniCPM V 4.7 1B
MiniCPM · WhiskyAKM
Params1.01B
Context256K
Smallest0.6 GB
ReasoningTool useVision
SP
Spark X2.5 4B
Spark · WhiskyAKM
Params4.11B
Context1024K
Smallest2.4 GB
ReasoningTool use
GE
Gemma 4 26B A4B it qat
Gemma · WhiskyAKM
Params25.2B
Context256K
Q4_K_M16.8 GB
Vision
GE
Gemma 4 E4B It Qat Q4_0 Unquantized
Gemma · WhiskyAKM
Params7.46B
Context128K
Q4_K_M5.3 GB
ReasoningTool useVision
MI
MiniCPM5 2B NVFP4
MiniCPM · WhiskyAKM
Params2.52B
Context128K
Smallest0.2 GB
ReasoningTool use
HO
K2-Horizon-MoVA-36B-A4B
Horizon · WhiskyAKM
Params37.4B
Context512K
Smallest21.3 GB
ReasoningTool use
SP
Spark X2.5 4B
Spark · WhiskyAKM
Params4.11B
Context1024K
Q4_K_M2.6 GB
ReasoningTool use
HO
K2-Horizon-7B
Horizon · WhiskyAKM
Params9B
Context512K
Smallest5.3 GB
ReasoningTool use
NE
Nemotron 3.5 Lightning 30B A3B NVFP4
Nemotron · WhiskyAKM
Params32.9B
Context1024K
Smallest0.7 GB
ReasoningTool use
QW
Qwen3.6 35B A3B
Qwen · WhiskyAKM
Params35.5B
Context256K
Smallest20.2 GB
ReasoningTool useVision
LF
LFM2.5 2.6B
Lfm · WhiskyAKM
Params2.7B
Context128K
Q4_K_M1.7 GB
ReasoningTool use
HO
K2-Horizon-3.7B
Horizon · WhiskyAKM
Params5.06B
Context512K
Smallest3.0 GB
ReasoningTool use

About WhiskyAKM on Hugging Face

WhiskyAKM has published 35 GGUF models that ggufmodels.com tracks, across families such as Gemma, Qwen and MiniCPM. The most downloaded are LFM2.5 24B A2B, Gemma 4 E2B it qat NVFP4 and Gemma 4 26B A4B it qat NVFP4.

The quants WhiskyAKM publishes most often are BF16 (15), Q8_0 (12), Q6_K (12) and Q5_K_S (12). Their Q4_K_M files range from 0.2 to 77.0 GB. Parameter counts run from 1.01B to 177B, and the longest trained context window is 1024K tokens.

Frequently asked questions

What GGUF models has WhiskyAKM published?

35 models are listed here, mainly from the Gemma, Qwen and MiniCPM families. The most downloaded is LFM2.5 24B A2B.

Which quants does WhiskyAKM publish?

The most common are BF16 (15 models), Q8_0 (12 models), Q6_K (12 models) and Q5_K_S (12 models).

Keep browsing