QW

Qwen3.5 4B NekoCOT

QwenGGUF by starryhoshiUpdated 2026-10-04
Download on Hugging Face
Parameters4.33B
Context256K
Layers33
Architectureqwen35
Smallest quant2.8 GB
Quick answer

Qwen3.5 4B NekoCOT needs about 4.4 GB of memory at Q4_K_M (a 2.8 GB file) with an 8K context window, so it fits on a 6 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 4.4 GB.

Qwen3.5 4B NekoCOT – a cat‑girl role‑play LLM with reasoning and vision support

Qwen3.5 4B NekoCOT is a 4.33 billion‑parameter model fine‑tuned with QLoRA for cat‑girl role‑play. It retains the Qwen3.5 vision encoder and adds a reasoning, role‑play, tools and vision adapter. The GGUF file is 2.78 GB in Q4_K_M quant.

Good at

  • Reasoning with "think before reply" style
  • Cat‑girl role‑play with emotional support
  • Vision input via base Qwen3.5 encoder

Limits

  • Requires 8 GB+ VRAM for 4‑bit inference
  • CC‑BY‑NC 4.0 license prohibits commercial use
  • No separate vision quant – mmproj must be loaded for image input

Best for

  • Chatting in a cat‑girl persona
  • Providing emotional companionship
  • Answering image‑based queries with cat‑girl flair

Who it’s for. Developers and hobbyists who want a local LLM with a strong role‑play personality and built‑in vision. The model is ideal for personal projects, educational demos, or non‑commercial creative applications.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M2.8 GB 4.4 GB
Fits Get file →

For 8 GB VRAM, use the Q4_K_M quantized file (2.78 GB). It needs about 4.4 GB of VRAM at 8 k context, leaving room for KV cache. If you have less VRAM, the model cannot run in 4‑bit mode. No other quants are provided in the file list.

What GPU do you need for Qwen3.5 4B NekoCOT?

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: the best quant that fits is Q4_K_M (2.8 GB file, about 4.4 GB in use).

Which Qwen3.5 4B NekoCOT quant should you choose?

Q4_K_M is the usual starting point for Qwen3.5 4B NekoCOT: it balances file size and quality well. Here it is a 2.8 GB file.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters4.33B
Layers33
Trained context262,144 tokens
KV cache (fp16)132 KB per token

Source

Quantized bystarryhoshi
FamilyQwen
Downloads (HF)934
Last updated2026-10-04

Frequently asked questions about Qwen3.5 4B NekoCOT

How much VRAM does Qwen3.5 4B NekoCOT need?

About 4.4 GB at Q4_K_M with an 8K context window. That is the 2.8 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 Qwen3.5 4B NekoCOT on an 8 GB GPU?

Yes, with Q4_K_M: the 2.8 GB file needs about 4.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Qwen3.5 4B NekoCOT GGUF?

It depends on the quant: Q4_K_M is 2.8 GB.

What context length does Qwen3.5 4B NekoCOT support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.4 GB at 32K.

Where can I download Qwen3.5 4B NekoCOT GGUF?

The files are on Hugging Face in the starryhoshi/Qwen3.5-4B-NekoCOT-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 Qwen3.5 4B NekoCOT use?

Hugging Face lists the license as cc-by-nc-4.0. 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 Qwen3.5 4B NekoCOT on 8 GB VRAM?

Yes, the Q4_K_M quantized file is 2.78 GB and requires about 4.4 GB of VRAM at 8 k context, which fits on an 8 GB GPU.

Does the model support image input?

Yes, the base Qwen3.5 vision encoder is included. Load the mmproj file to enable image or video input.

Is the model free for commercial use?

No. The CC‑BY‑NC 4.0 license prohibits commercial use of the fine‑tuned model.

Similar models

Browse all →

More Qwen models

All Qwen →

Other small models

All small models →