Qwen3.8 27B Heretic Uncensored Q8 0 needs about 31.7 GB of memory at Q8_0 (a 29.0 GB file) with an 8K context window, so it fits on a 32 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.
Qwen3.8‑27B Heretic Uncensored Q8 0 – 27.3B param dense vision‑language model
Qwen3.8‑27B Heretic Uncensored Q8 0 is a 27.3B parameter dense model with native vision‑language support. It is quantized to Q8_0 and runs on the qwen35 architecture. The model is licensed under apache‑2.0 and supports a 262144 token context.
Good at
- Vision‑language tasks
- Long‑horizon agentic reasoning
- Coding and professional work
Limits
- Requires ~31.7 GB VRAM at 8k context
- No built‑in safety alignment – uncensored behavior
- No external vision caveats beyond native support
Best for
- Deploying on high‑VRAM GPUs
- Research on multimodal reasoning
- Large‑scale coding assistance
Who it’s for. Engineers and researchers needing a dense, vision‑enabled LLM with a large context window. The model is suitable for environments that can provide the necessary VRAM and comply with the apache‑2.0 license.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q8_0 | 29.0 GB | 31.7 GB | Too big | Get file → |
The only available quant is Q8_0, 29.05 GB file size. It requires approximately 31.7 GB VRAM at 8k context. For 32 GB or larger GPUs, Q8_0 is suitable. Smaller GPUs cannot run this quant safely.
What GPU do you need for Qwen3.8 27B Heretic Uncensored Q8 0?
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: the best quant that fits is Q8_0 (29.0 GB file, about 31.7 GB in use).
Which Qwen3.8 27B Heretic Uncensored Q8 0 quant should you choose?
Q8_0 (29.0 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qwen3.8 27B Heretic Uncensored Q8 0
How much VRAM does Qwen3.8 27B Heretic Uncensored Q8 0 need?
About 31.7 GB at Q8_0 with an 8K context window. That is the 29.0 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 27B Heretic Uncensored Q8 0 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 31.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.8 27B Heretic Uncensored Q8 0 GGUF?
It depends on the quant: Q8_0 is 29.0 GB.
What context length does Qwen3.8 27B Heretic Uncensored Q8 0 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.2 GB at 8K and 8.7 GB at 32K.
Where can I download Qwen3.8 27B Heretic Uncensored Q8 0 GGUF?
The files are on Hugging Face in the mlasli/Qwen3.8-27B-Heretic-Uncensored-Q8_0-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 27B Heretic Uncensored Q8 0 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?
No. The Q8_0 quant requires about 31.7 GB VRAM at 8k context, so 8 GB is insufficient.
Does it support vision tasks?
Yes. It is a native vision‑language model that understands images and videos.
Can I use it with llama.cpp?
Yes, load it with llama.cpp using arch `qwen35` and enable MTP speculative decoding.