Qwen3.8 2B Uncensored needs about 2.2 GB of memory at Q4_K_M (a 1.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.
Qwen3.8 2B Uncensored – a small, text‑only GGUF model
Qwen3.8 2B Uncensored is a 1.9‑billion‑parameter causal language model based on the Qwen3.5 architecture. It is released as a Q4_K_M quantized GGUF file and is designed for CPU‑only inference on low‑end devices. The model has no vision support and is uncensored, allowing it to generate unrestricted content.
Good at
- Reasoning and general text generation
- Fast CPU inference with low memory
- Large context of 262,144 tokens
Limits
- No vision or multimodal support
- Uncensored content may generate harmful text
- Requires Apache‑2.0 license compliance
Best for
- Prototyping language models on SBCs
- Fine‑tuning for research tasks
- Generating long‑form text with large context
Who it’s for. Developers and researchers who need a lightweight, text‑only model that can run on CPUs with limited VRAM. The model is suitable for experimentation and custom fine‑tuning, but not for production use that requires content filtering.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 1.3 GB | 2.2 GB | Fits | Get file → |
The Q4_K_M quantized file is 1.27 GB and requires about 2.2 GB of RAM at an 8 k context. For 8 GB VRAM devices, this model fits comfortably. If memory is tighter, consider a higher‑quantization variant, but only Q4_K_M is available in the file list.
What GPU do you need for Qwen3.8 2B Uncensored?
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: the best quant that fits is Q4_K_M (1.3 GB file, about 2.2 GB in use).
Which Qwen3.8 2B Uncensored quant should you choose?
Q4_K_M is the usual starting point for Qwen3.8 2B Uncensored: it balances file size and quality well. Here it is a 1.3 GB file.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qwen3.8 2B Uncensored
How much VRAM does Qwen3.8 2B Uncensored need?
About 2.2 GB at Q4_K_M with an 8K context window. That is the 1.3 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 Qwen3.8 2B Uncensored on an 8 GB GPU?
Yes, with Q4_K_M: the 1.3 GB file needs about 2.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Qwen3.8 2B Uncensored GGUF?
It depends on the quant: Q4_K_M is 1.3 GB.
What context length does Qwen3.8 2B Uncensored support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.6 GB at 32K.
Where can I download Qwen3.8 2B Uncensored GGUF?
The files are on Hugging Face in the atakhadivi/Qwen3.8-2B-Uncensored-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 Qwen3.8 2B Uncensored 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 Qwen3.8 2B Uncensored on 8 GB VRAM?
Yes. The Q4_K_M file uses 1.27 GB of storage and needs roughly 2.2 GB of RAM at 8 k context, well within an 8 GB VRAM budget.
Does this model support vision inputs?
No. The GGUF file is text‑only and does not include a vision projector or MTP head.
Is the model safe for production use?
The model is uncensored and may generate harmful content. It is intended for research or personal use only.