KI

Kimi-K3

KimiGGUF by 6blockUpdated 2026-08-03
Download on Hugging Face
Parameters2780B
Context1024K
Layers93
Architecturekimi-k3
Smallest quant540.4 GB
Quick answer

Kimi-K3 needs about 636.0 GB of memory at IQ1_S (a 540.4 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 2 quants, from 540.4 GB (IQ1_S) to 579.5 GB (IQ1_S). Longer context needs more: at 32K tokens the KV cache alone adds about 380.3 GB.

Kimi-K3: 2.8T‑parameter multimodal LLM with 1‑million‑token context

Kimi-K3 is a 2.8T‑parameter multimodal model built on Kimi Delta Attention and Attention Residuals. It supports text, images, and video, and offers a 1‑million‑token context window. The model is a mixture‑of‑experts with 896 experts, 16 active per token.

Good at

  • Long‑horizon coding
  • Agentic knowledge work
  • Native multimodality

Limits

  • No vision support in quantized GGUFs
  • Requires large VRAM for full model
  • License is other

Best for

  • Research on multimodal LLMs
  • Long‑context document analysis
  • Tool‑integrated coding sessions

Who it’s for. Researchers and developers needing a large multimodal model with extensive context. The model is open‑weight but requires significant compute resources.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S540.4 GB (11 parts) 636.0 GB
Too big Get file →
IQ1_S579.5 GB (34 parts) 675.1 GB
Too big Get file →

For 8×H100 + 2TB RAM nodes, the IQ1_S quantization fits. The IQ1_S‑XXS‑832e tier reduces size to under 512 GiB by pruning experts. Choose IQ1_S‑XXS‑832e for lower memory budgets, IQ1_S for maximum quality.

What GPU do you need for Kimi-K3?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) 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.

Which Kimi-K3 quant should you choose?

IQ1_S (540.4 GB) is the middle option in this repository. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturekimi-k3
Parameters2780B (approximate, mixture of experts)
Layers93
Trained context1,048,576 tokens
KV cache (fp16)11334 KB per token

Source

Quantized by6block
FamilyKimi
Licenseother
Downloads (HF)1,115
Last updated2026-08-03

Frequently asked questions about Kimi-K3

How much VRAM does Kimi-K3 need?

About 636.0 GB at IQ1_S with an 8K context window. That is the 540.4 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 Kimi-K3 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 636.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Kimi-K3 GGUF?

It depends on the quant: IQ1_S is 540.4 GB, IQ1_S is 579.5 GB.

What context length does Kimi-K3 support?

The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 95.1 GB at 8K and 380.3 GB at 32K.

Where can I download Kimi-K3 GGUF?

The files are on Hugging Face in the 6block/Kimi-K3-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 Kimi-K3 use?

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

Can I run Kimi-K3 on 8 GB VRAM?

No. The model requires a large VRAM setup; the smallest quantized tier still needs a multi‑GPU node.

Does the GGUF version support vision?

The provided GGUF files are quantized for text; vision support is not included in these tiers.

What is the context length of Kimi-K3?

The model supports a 1,048,576‑token context window.

Other GGUFs of Kimi-K3

Similar models

Browse all →

More from 6block

All by 6block →

More Kimi models

All Kimi →

Other large models

All large models →