Mistral-Nemo-2407-12B-Thinking-Claude-Gemini-GPT5.2-Uncensored-HERETIC
from mistralai/Mistral-Nemo-Instruct-2407
Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus is a Heretic-ablated variant of mistralai/Mistral-Nemo-Instruct-2407, published on Hugging Face by DavidAU. It is 12B parameters and 24GB VRAM class class. It is distributed across 13 repositories in Transformers, GGUF, GGUF (imatrix) and MLX formats, totalling 5.2K downloads.
How Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus was modified, and what that implies.
Heretic
Automated directional ablation via the Heretic toolchain, which searches for the refusal direction and applies it with a KL-divergence budget so general capability is preserved.
Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.
From the publisher’s model card
“WARNING "HERETIC" version: Unlocked. UNFILTERED. NSFW. Vivid prose. INTENSE. Visceral Details. Light to R-18 HORROR. Swearing. UNCENSORED... humor, romance, fun... and UNFILTERED TRUTH.”
Commands are templates — confirm the exact repository and quant file before use.
hf download mradermacher/Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus-GGUF --local-dir ./mistral-nemo-inst-2407-12b-thinking-uncensored-heretic-hi-claude-opus
llama-cli -m ./mistral-nemo-inst-2407-12b-thinking-uncensored-heretic-hi-claude-opus/<file>.gguf -p "..."from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "DavidAU/Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vllm serve DavidAU/Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus --trust-remote-codeEvery published repository of Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus, including quantised re-releases by other authors.
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.
Metadata for Mistral-Nemo-Inst-2407-12B-Thinking-Uncensored-HERETIC-HI-Claude-Opus is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with DavidAU, and does not independently verify publisher claims. See responsible use.