Gemma4-12B-QAT-Uncensored-HauhauCS-Balanced
12BGemmauncensored
from google/gemma-4-12B-it
↓ 428.4K♥ 3325 repos
Gemma-4-12B-OBLITERATED
Gemma-4-12B-OBLITERATED is a abliterated and uncensored variant of google/gemma-4-12B-it, published on Hugging Face by threlfall-hax. It is 12B parameters, 24GB VRAM class class and gemma licence. It is distributed across 25 repositories in Transformers, GGUF, GGUF (imatrix) and MLX formats, totalling 37.7K downloads.
Specification
- Model name
- Gemma-4-12B-OBLITERATED
- Source repository
- threlfall-hax/Gemma-4-12B-OBLITERATED
- Base model
- google/gemma-4-12B-it
- Publisher
- threlfall-hax
- Parameters
- 12B
- Hardware class
- 10B – 25B — 24GB VRAM class
- Licence
- gemma
- Task
- Unspecified
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF, GGUF (imatrix), MLX
- First indexed
- 5 Jun 2026
- Last updated
- 4 Sept 2026
Ablation technique
How Gemma-4-12B-OBLITERATED was modified, and what that implies.
Abliteration + uncensored fine-tune
Directional ablation combined with additional fine-tuning on unfiltered data, so behaviour diverges from the base model beyond refusal removal alone.
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
“An abliterated (uncensored) version of google/gemma-4-12b-it with refusal behavior removed.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download OBLITERATUS/Gemma-4-12B-OBLITERATED --local-dir ./gemma-4-12b-obliterated
llama-cli -m ./gemma-4-12b-obliterated/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "threlfall-hax/Gemma-4-12B-OBLITERATED"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve threlfall-hax/Gemma-4-12B-OBLITERATED --trust-remote-codeDownloads and variants (24)
Every published repository of Gemma-4-12B-OBLITERATED, including quantised re-releases by other authors.
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Gemma-4-12B-OBLITERATED will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
- Capability is inherited from google/gemma-4-12B-it, not added by ablation — at 12B parameters, expect roughly the base model’s competence, minus some instruction-following fidelity.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- This release adds fine-tuning on unfiltered data, so its behaviour diverges from the base model beyond refusal removal — differences you measure cannot be attributed to ablation alone.
- Declared licence is gemma, but the terms that bind you are google/gemma-4-12B-it’s — a re-release cannot grant more than its parent.
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 Gemma-4-12B-OBLITERATED is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with threlfall-hax, and does not independently verify publisher claims. See responsible use.