Qwen3.6 35B A3B Uncensored Heretic needs about 12.8 GB of memory at IQ2_M (a 11.7 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 2 quants, from 11.7 GB (IQ2_M) to 12.3 GB (IQ2_M). Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.
Qwen3.6 35B A3B Uncensored Heretic – a large, uncensored LLM with reduced refusals
This model is a 34.7B‑parameter Qwen variant that has been modified with Heretic abliteration to lower content refusals while keeping reasoning, coding, and language skills. It is quantized to IQ2_M for efficient local inference. The model supports vision, reasoning, and tool use.
Good at
- Reasoning and problem solving
- Tool usage and code generation
- Vision tasks
Limits
- Requires ~13 GB VRAM at 8 k context
- No official vision API support
- Apache‑2.0 license only
Best for
- Local inference on high‑VRAM GPUs
- Developers needing uncensored responses
- Researchers testing refusal reduction
Who it’s for. Engineers and researchers who need a large, uncensored LLM that can run locally on GPUs with at least 13 GB of VRAM. The model is suitable for applications that require reasoning, coding, and vision capabilities without external API calls.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_M | 11.7 GB | 12.8 GB | Too big | Get file → | |
| IQ2_M | 12.3 GB | 13.5 GB | Too big | Get file → |
For a 12 GB VRAM budget, use the 11.66 GB IQ2_M file; it estimates 12.8 GB memory at 8 k context. If you have 13 GB or more, the 12.33 GB file is available and estimates 13.5 GB memory. Both are dynamic IQ2_M quantizations from llama.cpp.
What GPU do you need for Qwen3.6 35B A3B Uncensored Heretic?
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: the best quant that fits is IQ2_M (12.3 GB file, about 13.5 GB in use).
Which Qwen3.6 35B A3B Uncensored Heretic quant should you choose?
IQ2_M (11.7 GB) is the middle option in this repository. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about Qwen3.6 35B A3B Uncensored Heretic
How much VRAM does Qwen3.6 35B A3B Uncensored Heretic need?
About 12.8 GB at IQ2_M with an 8K context window. That is the 11.7 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 Qwen3.6 35B A3B Uncensored Heretic on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 12.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.6 35B A3B Uncensored Heretic GGUF?
It depends on the quant: IQ2_M is 11.7 GB, IQ2_M is 12.3 GB.
What context length does Qwen3.6 35B A3B Uncensored Heretic support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.
Where can I download Qwen3.6 35B A3B Uncensored Heretic GGUF?
The files are on Hugging Face in the BlueBackup/Qwen3.6-35B-A3B-uncensored-heretic-IQ2_M repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Qwen3.6 35B A3B Uncensored Heretic 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 model needs at least 12.8 GB VRAM at 8 k context, so 8 GB is insufficient.
Does this model support vision?
Yes, it has detected vision abilities, but no dedicated vision API is provided.
What license applies?
The model is released under the Apache‑2.0 license.