QW

Qwen3.5 4B UNCENSORED NONREASONING

QwenGGUF by heavylildudeUpdated 2026-04-20
Download on Hugging Face
Parameters4.21B
Context256K
Layers32
Architectureqwen35
Smallest quant2.7 GB
Quick answer

Qwen3.5 4B UNCENSORED NONREASONING needs about 4.3 GB of memory at Q4_K_M (a 2.7 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 4.3 GB.

Qwen3.5 4B UNCENSORED NONREASONING – fast, direct responses

Qwen3.5 4B UNCENSORED NONREASONING is a stripped‑down variant of Qwen 3.5 4B. It removes reasoning traces and special output formatting for quicker, more direct answers. The model supports multimodal vision and works with llama.cpp and related tools.

Good at

  • Fast local inference
  • Direct response generation
  • Multimodal vision support

Limits

  • No chain‑of‑thought reasoning
  • No special prompt flags
  • Requires 8.3GB VRAM for 8k context

Best for

  • Chatting
  • Coding assistance
  • Local AI workflows

Who it’s for. Developers and users who need quick, direct answers without the overhead of reasoning chains. Ideal for local inference environments that prioritize speed.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M2.7 GB 4.3 GB
Fits Get file →

The Q4_K_M quant is 2.71GB and needs about 4.3GB of VRAM at 8k context. This is suitable for systems with 8GB or more VRAM. For tighter budgets, a higher‑quantization level is not available in the current file list.

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

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 (2.7 GB file, about 4.3 GB in use).

Which Qwen3.5 4B UNCENSORED NONREASONING quant should you choose?

Q4_K_M is the usual starting point for Qwen3.5 4B UNCENSORED NONREASONING: it balances file size and quality well. Here it is a 2.7 GB file.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters4.21B
Layers32
Trained context262,144 tokens
KV cache (fp16)128 KB per token

Source

Quantized byheavylildude
FamilyQwen
Licenseapache-2.0
Downloads (HF)580
Last updated2026-04-20

Frequently asked questions about Qwen3.5 4B UNCENSORED NONREASONING

How much VRAM does Qwen3.5 4B UNCENSORED NONREASONING need?

About 4.3 GB at Q4_K_M with an 8K context window. That is the 2.7 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 Qwen3.5 4B UNCENSORED NONREASONING on an 8 GB GPU?

Yes, with Q4_K_M: the 2.7 GB file needs about 4.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

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

It depends on the quant: Q4_K_M is 2.7 GB.

What context length does Qwen3.5 4B UNCENSORED NONREASONING 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 4.3 GB at 32K.

Where can I download Qwen3.5 4B UNCENSORED NONREASONING GGUF?

The files are on Hugging Face in the heavylildude/Qwen3.5-4b-Uncensored-NonReasoning 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 UNCENSORED NONREASONING use?

Hugging Face lists the license as apache-2.0. 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 this model on 8 GB VRAM?

Yes, the Q4_K_M quant uses about 4.3GB of VRAM at 8k context, leaving headroom for other processes.

Does it support vision?

Yes, the model retains multimodal/vision support as described in the repo.

Is there a license restriction?

The model is released under the apache-2.0 license, allowing free use and modification.

Similar models

Browse all →

More from heavylildude

All by heavylildude →

More Qwen models

All Qwen →

Other small models

All small models →