Gemma4-12B-QAT-Uncensored-HauhauCS-Balanced
12BGemmauncensored
from google/gemma-4-12B-it
↓ 428.4K♥ 3325 repos
gemma-4-12B-it-abliterix
gemma-4-12B-it-abliterix is a abliterated and uncensored variant of wangzhang/gemma-4-12B-it-abliterix, published on Hugging Face by wangzhang. It is 12B parameters, 24GB VRAM class class and apache-2.0 licence. It is distributed across 5 repositories in Transformers, GGUF and GGUF (imatrix) formats, totalling 5K downloads.
Specification
- Model name
- gemma-4-12B-it-abliterix
- Source repository
- wangzhang/gemma-4-12B-it-abliterix
- Base model
- wangzhang/gemma-4-12B-it-abliterix
- Publisher
- wangzhang
- Parameters
- 12B
- Hardware class
- 10B – 25B — 24GB VRAM class
- Licence
- apache-2.0
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF, GGUF (imatrix)
- First indexed
- 23 Jun 2026
- Last updated
- 2 Sept 2026
Ablation technique
How gemma-4-12B-it-abliterix 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 uncensored, refusal-suppressed version of google/gemma-4-12B-it, produced by directional ablation (no fine-tuning, no new data) with abliterix.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download mradermacher/gemma-4-12B-it-abliterix-i1-GGUF --local-dir ./gemma-4-12b-it-abliterix
llama-cli -m ./gemma-4-12b-it-abliterix/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "wangzhang/gemma-4-12B-it-abliterix"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve wangzhang/gemma-4-12B-it-abliterix --trust-remote-codeDownloads and variants (4)
Every published repository of gemma-4-12B-it-abliterix, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| mradermacher/gemma-4-12B-it-abliterix-i1-GGUF | GGUF (imatrix) | 3.7K |
| mradermacher/gemma-4-12B-it-abliterix-GGUF | GGUF | 689 |
| wangzhang/gemma-4-12B-it-abliterixsource | Transformers | 598 |
| 0xA50C1A1/gemma-4-12B-it-abliterix-gguf | GGUF | 25 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- gemma-4-12B-it-abliterix 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 wangzhang/gemma-4-12B-it-abliterix, 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 apache-2.0, but the terms that bind you are wangzhang/gemma-4-12B-it-abliterix’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-it-abliterix is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with wangzhang, and does not independently verify publisher claims. See responsible use.