KA

Kandinsky 6.0 Pro 5s GGUFs

KandinskyGGUF by realrebelaiUpdated 2026-10-08
Download on Hugging Face
Parameters—
Context—
Layers—
Architecturewan
Smallest quant20.4 GB
Quick answer

Kandinsky 6.0 Pro 5s GGUFs needs about 23.3 GB of memory at Q3_K_S (a 20.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 2 quants, from 20.4 GB (Q3_K_S) to 23.7 GB (Q4_K_S).

Kandinsky 6.0 Pro 5s GGUFs – low‑VRAM image generation

Kandinsky 6.0 Pro 5s GGUFs provide two quantized variants for ComfyUI. The Q3_K_S and Q4_K_S files reduce model size and memory usage while maintaining the 50‑step Pro workflow. The model is released under the MIT license.

Good at

  • Low‑VRAM inference
  • ComfyUI compatibility
  • Audio/video conditioning

Limits

  • Experimental, not fully tested
  • No vision support beyond image conditioning
  • Requires specific ComfyUI forks

Best for

  • Desktop image generation
  • Low‑resource inference
  • ComfyUI workflow prototyping

Who it’s for. Users who need a lightweight version of Kandinsky 6.0 Pro 5s for ComfyUI and can work with experimental builds. The model is suitable for developers and researchers testing new quantizations.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_S20.4 GB 23.3 GB
Too big Get file →
Q4_K_S23.7 GB 27.1 GB
Too big Get file →

For 8‑GB VRAM, use Q3_K_S (file size 20.35 GB, estimated 23.3 GB memory at 8k context). If more memory is available, Q4_K_S (23.73 GB, 27.1 GB memory) offers slightly higher precision.

What GPU do you need for Kandinsky 6.0 Pro 5s GGUFs?

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: the best quant that fits is Q3_K_S (20.4 GB file, about 23.3 GB in use).
  • 32 GB VRAM: the best quant that fits is Q4_K_S (23.7 GB file, about 27.1 GB in use).

Which Kandinsky 6.0 Pro 5s GGUFs quant should you choose?

Q3_K_S (20.4 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturewan

Source

Quantized byrealrebelai
FamilyKandinsky
Licensemit
Downloads (HF)5,351
Last updated2026-10-08

Frequently asked questions about Kandinsky 6.0 Pro 5s GGUFs

How much VRAM does Kandinsky 6.0 Pro 5s GGUFs need?

About 23.3 GB at Q3_K_S with an 8K context window. That is the 20.4 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 Kandinsky 6.0 Pro 5s GGUFs on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 23.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Kandinsky 6.0 Pro 5s GGUFs GGUF?

It depends on the quant: Q3_K_S is 20.4 GB, Q4_K_S is 23.7 GB.

Where can I download Kandinsky 6.0 Pro 5s GGUFs GGUF?

The files are on Hugging Face in the realrebelai/Kandinsky-6.0-Pro-5s_GGUFs repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Kandinsky 6.0 Pro 5s GGUFs 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.

Can I run Q3_K_S on 8 GB VRAM?

The estimated memory usage is 23.3 GB at 8k context, so 8 GB VRAM is insufficient. You would need at least 24 GB VRAM to run Q3_K_S safely.

What license does this model use?

The model is released under the MIT license, allowing free use and modification.

Do I need special ComfyUI nodes for GGUF?

Yes. GGUF users must install the Kandinsky6_GGUF node set and the ComfyUI-GGUF-Rebel fork to ensure compatibility.

More from realrebelai

All by realrebelai →

More Kandinsky models

All Kandinsky →