Huihui Qwen3.8 27B Coder390 EfficientThink abliterated needs about 14.1 GB of memory at its main file (a 12.1 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 7 quants, from 0.0 GB to 54.7 GB (BF16).
Huihui Qwen3.8 27B Coder390 EfficientThink abliterated – an uncensored, ablated Qwen model
This model is an ablated version of Qwen3.8‑27B‑Coder390‑EfficientThink. It removes refusals by ablating layers 2 to 52. The GGUF package includes several quantization tiers, including BF16 and unknown quantizations.
Good at
- Uncensored text generation
- Ablation of refusal behavior
- Compatibility with llama.cpp
Limits
- Requires large VRAM for full quantization
- No vision support mentioned
- License is apache‑2.0
Best for
- Local LLM deployment
- Testing ablation techniques
- Research on refusal removal
Who it’s for. Developers and researchers who need an uncensored Qwen model for local inference. The model is suitable for those who can manage the memory requirements and want to experiment with ablation.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.0 GB | 0.5 GB | Fits | Get file → | |
| — | 0.0 GB | 0.5 GB | Fits | Get file → | |
| — | 0.0 GB | 0.5 GB | Fits | Get file → | |
| — | 12.1 GB | 14.1 GB | Too big | Get file → | |
| — | 12.1 GB | 14.1 GB | Too big | Get file → | |
| — | 12.1 GB | 14.1 GB | Too big | Get file → | |
| BF16 | 54.7 GB | 61.7 GB | Too big | Get file → |
For 8 GB VRAM, use the 0.03 GB unknown quant files; they estimate 0.5 GB memory at 8 k context. For 16 GB VRAM, the 12.14 GB unknown quant files use about 14.1 GB memory. For larger GPUs, the BF16 54.66 GB file uses about 61.7 GB memory.
What GPU do you need for Huihui Qwen3.8 27B Coder390 EfficientThink abliterated?
These figures add the file size, an fp16 KV cache at an 8K context window 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 the main file (0.0 GB file, about 0.5 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 0.5 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 0.5 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 0.5 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (12.1 GB file, about 14.1 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (12.1 GB file, about 14.1 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (12.1 GB file, about 14.1 GB in use).
- 48 GB VRAM: the best quant that fits is the main file (12.1 GB file, about 14.1 GB in use).
Which Huihui Qwen3.8 27B Coder390 EfficientThink abliterated quant should you choose?
If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about Huihui Qwen3.8 27B Coder390 EfficientThink abliterated
How much VRAM does Huihui Qwen3.8 27B Coder390 EfficientThink abliterated need?
About 14.1 GB at its main file with an 8K context window. That is the 12.1 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 Huihui Qwen3.8 27B Coder390 EfficientThink abliterated on an 8 GB GPU?
Yes, with the smaller files: the 0.0 GB file needs about 0.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Huihui Qwen3.8 27B Coder390 EfficientThink abliterated GGUF?
It depends on the quant: file is 0.0 GB, file is 0.0 GB, file is 0.0 GB, file is 12.1 GB, file is 12.1 GB, file is 12.1 GB and 1 more.
Where can I download Huihui Qwen3.8 27B Coder390 EfficientThink abliterated GGUF?
The files are on Hugging Face in the huihui-ai/Huihui-Qwen3.8-27B-Coder390-EfficientThink-abliterated-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 Huihui Qwen3.8 27B Coder390 EfficientThink abliterated 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, use the 0.03 GB unknown quant files; they estimate 0.5 GB memory at 8 k context.
What quantization is best for limited memory?
The 12.14 GB unknown quant files use about 14.1 GB memory, suitable for 16 GB VRAM setups.
Is the model multimodal?
The parent model is fully multimodal, but the ablated version’s multimodal support is not specified.