DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora needs about 0.6 GB of memory at its main file (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 8 quants, from 0.0 GB to 0.2 GB.
DeepSeek V4 Flash 0731 Heretic Abiliterated v2 LoRA GGUF model
DeepSeek V4 Flash 0731 Heretic Abiliterated v2 is a LoRA adapter set for the DeepSeek V4 Flash 0731 base model. It removes the base model’s refusal behavior and is tuned for decensoring. The adapters are lightweight, around 50 MB each, and can be applied to any GGUF quant of the base model.
Good at
- Decensoring and reducing refusals
- Maintaining base model performance
- LoRA-based lightweight download
Limits
- No vision support
- Requires base model separately
- No safety guardrails
Best for
- Fine‑tuning for content‑sensitive applications
- Deploying on low‑memory devices
- Research on refusal mitigation
Who it’s for. Developers who need a decensored version of DeepSeek V4 Flash 0731 and have the base model available. The LoRA adapters are lossless and can be merged with any GGUF quant of the base model.
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 → | |
| — | 0.1 GB | 0.6 GB | Fits | Get file → | |
| — | 0.1 GB | 0.6 GB | Fits | Get file → | |
| — | 0.1 GB | 0.6 GB | Fits | Get file → | |
| — | 0.2 GB | 0.7 GB | Fits | Get file → | |
| — | 0.2 GB | 0.7 GB | Fits | Get file → | |
| — | 0.2 GB | 0.7 GB | Fits | Get file → |
For 8 GB VRAM, use the 0.08 GB or 0.09 GB adapters; they need about 0.6 GB memory at 8 k context. For 12 GB VRAM, the 0.16 GB or 0.18 GB adapters are suitable, requiring roughly 0.7 GB. The 0.2 GB adapter is best for 16 GB VRAM, with an estimate of 0.7 GB memory. Estimates are approximate.
What GPU do you need for DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora?
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 (0.2 GB file, about 0.7 GB in use).
Which DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora 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
Source
Frequently asked questions about DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora
How much VRAM does DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora need?
About 0.6 GB at its main file with an 8K context window. That is the 0.1 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 DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora on an 8 GB GPU?
Yes, with the smaller files: the 0.2 GB file needs about 0.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora GGUF?
It depends on the quant: file is 0.0 GB, file is 0.0 GB, file is 0.1 GB, file is 0.1 GB, file is 0.1 GB, file is 0.2 GB and 2 more.
Where can I download DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora GGUF?
The files are on Hugging Face in the MorinoNushi/DeepSeek-V4-Flash-0731-heretic-abliterated-v2-GGUF-lora repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does DeepSeek V4 Flash 0731 heretic abliterated v2 GGUF lora use?
Hugging Face lists the license as mit. 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 the 0.08 GB adapter on a GPU with 8 GB VRAM?
Yes, the 0.08 GB adapter requires about 0.6 GB memory at 8 k context, which fits within 8 GB VRAM.
Do I need the base model to use these adapters?
Yes, the adapters are applied to any GGUF quant of the DeepSeek V4 Flash 0731 base model.
Are there safety guardrails in the adapted model?
No, the adapters suppress the base model’s refusal behavior, so the resulting model has reduced safety guardrails.