Qwen3.8 27B Uncensored Heretic T10 BF16 needs about 5.8 GB of memory at its main file (a 3.2 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 2 quants, from 3.2 GB to 28.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.
Qwen3.8 27B Uncensored Heretic T10 BF16 – dense vision‑language LLM
Qwen3.8 27B Uncensored Heretic T10 BF16 is a 26.9 billion‑parameter dense model built on the Qwen3.5 architecture. It supports reasoning, tools, and vision tasks and offers a 262144‑token context window. The model is released under the apache‑2.0 license.
Good at
- Vision‑language reasoning
- Tool‑augmented task execution
- Long‑horizon agentic planning
Limits
- No mixture of experts
- Limited to 262144 context tokens
- Requires BF16 or higher precision for full capability
Best for
- Complex multi‑step problem solving
- Interactive agent workflows
- Image‑and‑text data analysis
Who it’s for. Developers needing a compact, dense vision‑language model for local inference. Ideal for applications that require reasoning, tool use, and image understanding without large distributed setups.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 3.2 GB | 5.8 GB | Fits | Get file → | |
| Q8_0 | 28.6 GB | 31.3 GB | Too big | Get file → |
For 8 GB VRAM, use the Q8_0 quantized file (28.6 GB, ~31.3 GB memory at 8 k context). For 4 GB VRAM, the unknown quant file (3.16 GB, ~5.8 GB memory at 8 k context) is the only option. Memory estimates are approximate.
What GPU do you need for Qwen3.8 27B Uncensored Heretic T10 BF16?
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 the main file (3.2 GB file, about 5.8 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (3.2 GB file, about 5.8 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (3.2 GB file, about 5.8 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (3.2 GB file, about 5.8 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (3.2 GB file, about 5.8 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (28.6 GB file, about 31.3 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qwen3.8 27B Uncensored Heretic T10 BF16
How much VRAM does Qwen3.8 27B Uncensored Heretic T10 BF16 need?
About 5.8 GB at its main file with an 8K context window. That is the 3.2 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 Uncensored Heretic T10 BF16 on an 8 GB GPU?
Yes, with the smaller files: the 3.2 GB file needs about 5.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Qwen3.8 27B Uncensored Heretic T10 BF16 GGUF?
It depends on the quant: file is 3.2 GB, Q8_0 is 28.6 GB.
What context length does Qwen3.8 27B Uncensored Heretic T10 BF16 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 Uncensored Heretic T10 BF16 GGUF?
The files are on Hugging Face in the nerkyor/Qwen3.8-27B-Uncensored-Heretic-T10-BF16 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 Uncensored Heretic T10 BF16 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 Q8_0 quantized file requires about 31.3 GB memory at 8 k context, which exceeds 8 GB VRAM. Use the unknown quant file instead, which needs roughly 5.8 GB memory.
Does this model support vision tasks?
Yes, it includes vision capabilities as part of its detected abilities.
What license applies to this model?
The model is distributed under the apache‑2.0 license.