Offrail 27B Uncensored needs about 9.0 GB of memory at its main file (a 6.3 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 2 quants, from 6.3 GB to 7.0 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.
Offrail 27B Uncensored – a 27B ternary Qwen35 model with a Heretic LoRA graft.
Offrail 27B Uncensored is a 27‑billion‑parameter ternary model based on Ternary‑Bonsai‑2. It includes a Heretic LoRA baked into the ternary weights and a Qwen 3.8 27B multi‑token‑prediction head grafted back on. The model is released under Apache‑2.0 and is intended for research and safety studies.
Good at
- Large context handling up to 262,144 tokens
- Uncensored content generation
- Research on safety alignment
Limits
- No vision support
- Reduced safety alignment – may generate harmful content
- Requires PrismML llama.cpp fork for ternary kernels
Best for
- Safety research and red‑team testing
- Academic experiments with large context models
- Exploring uncensored language generation
Who it’s for. Researchers, safety engineers, and developers who need a large, uncensored ternary model for experimentation. The model is not suitable for public or end‑user deployments.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 6.3 GB | 9.0 GB | Too big | Get file → | |
| — | 7.0 GB | 9.7 GB | Too big | Get file → |
The 6.3 GB file needs about 9 GB VRAM at 8 k context, while the 7.01 GB file needs about 9.7 GB. For 8 GB VRAM use the 6.3 GB file; for 12 GB or more use the 7.01 GB file. Memory estimates are approximate.
What GPU do you need for Offrail 27B 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: 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: the best quant that fits is the main file (7.0 GB file, about 9.7 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Offrail 27B Uncensored
How much VRAM does Offrail 27B Uncensored need?
About 9.0 GB at its main file with an 8K context window. That is the 6.3 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 Offrail 27B Uncensored on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 9.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Offrail 27B Uncensored GGUF?
It depends on the quant: file is 6.3 GB, file is 7.0 GB.
What context length does Offrail 27B 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 2.2 GB at 8K and 8.7 GB at 32K.
Where can I download Offrail 27B Uncensored GGUF?
The files are on Hugging Face in the sudoingx/Offrail-27B-Uncensored repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Offrail 27B 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 Offrail 27B on an 8 GB GPU?
Yes, the 6.3 GB quantized file requires about 9 GB VRAM at 8 k context, which fits on an 8 GB GPU with some overhead.
Does this model support vision?
No, vision behavior has not been evaluated and the model does not include vision capabilities.
What license applies to this model?
The model is released under the Apache‑2.0 license, as stated in the parent model card.