KI

K3 Q4 ZUG

KimiGGUF by zugquantUpdated 2026-07-30
Download on Hugging Face
Parameters3470B
Context128K
Layers—
Architecturedeepseek2
Smallest quant2098.7 GB
Quick answer

K3 Q4 ZUG needs about 2351.0 GB of memory at its main file (a 2098.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds.

K3 Q4 ZUG is a 3470B-parameter Kimi model in GGUF format, quantized by zugquant. Its trained context length is 128K tokens. The repository has 1 quant file, from 2098.7 GB to 2098.7 GB.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—2098.7 GB (125 parts) 2351.0 GB
Too big Get file →

What GPU do you need for K3 Q4 ZUG?

These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Specifications

Architecture from the GGUF header

Architecturedeepseek2
Parameters3470B
Trained context131,072 tokens

Source

Quantized byzugquant
FamilyKimi
Licensemit
Downloads (HF)0
Last updated2026-07-30

Frequently asked questions about K3 Q4 ZUG

How much VRAM does K3 Q4 ZUG need?

About 2351.0 GB at its main file with an 8K context window. That is the 2098.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run K3 Q4 ZUG on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 2351.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of K3 Q4 ZUG GGUF?

It depends on the quant: file is 2098.7 GB.

What context length does K3 Q4 ZUG support?

The GGUF header lists a trained context length of 131,072 tokens (128K).

Where can I download K3 Q4 ZUG GGUF?

The files are on Hugging Face in the zugquant/K3-Q4-ZUG-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does K3 Q4 ZUG use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Other GGUFs of Kimi-K2-Base

Similar models

Browse all →

More Kimi models

All Kimi →

Other large models

All large models →