license: apache-2.0
library_name: llama.cpp
pipeline_tag: text-generation
base_model:
- mistralai/Mistral-Nemo-2407-12B-Thinking-Claude-Gemini-GPT5.2-Uncensored-HERETIC
language: - en
- fr
- de
- es
- it
- pt
- ru
- zh
- ja
tags: - uncensored
- heretic
- abliterated
- creative
- creative writing
- fiction writing
- plot generation
- sub-plot generation
- fiction writing
- story generation
- scene continue
- storytelling
- fiction story
- science fiction
- romance
- all genres
- story
- writing
- vivid prose
- vivid writing
- fiction
- roleplaying
- swearing
- mistral nemo
- horror
- context 1024-2048
Credits & License
This is a converted to *.gguf and quantized version of safetensors model. DavidAU/Mistral-Nemo-2407-12B-Thinking-Claude-Gemini-GPT5.2-Uncensored-HERETIC (And size of it converted from 24.5GB to 6.08GB)
So it can be used on CPU with 16GB of RAM (slow, Also can be run on Google Colab).
Problems:
- More stupid than original model
- If you are asking simple question without context (like "Hi!") the model will doomloop but if you ask more complex like "create a plot of non-fiction story" it will works normal.
- context window - 2048 minimum (can work with 1024 but)
My Contribution:
- Converted the safetensors (e.g. hf_model) to *.gguf
- Quantized to Q3_K_M
- Alarm: Quantized version can hallucinate.
- Recommended context window: 2048 tokens (or higher if hardware permits).
License:
This derivative work is distributed under the Apache 2.0 license, inherited from the original model.
Please refer to the original model card for full terms and disclaimer.
Donation address (if you want to see more quantized uncensored models - or in future create of more uncensored model):
1Kk32KvJVjcyTMZ44eos4eZtme1quTqFAg (BTC)