gemma-4-31B-it-abliterated
31BGemma
from google/gemma-4-31B-it
↓ 766.5K♥ 11845 repos
gemma-4-31B-it-uncensored
gemma-4-31B-it-uncensored is a abliterated and uncensored variant of google/gemma-4-31B-it, published on Hugging Face by TrevorJS. It is 31B parameters, 48GB VRAM class class and apache-2.0 licence. It is distributed across 36 repositories in GGUF, Transformers, MLX and GGUF (imatrix) formats, totalling 291K downloads.
Specification
- Model name
- gemma-4-31B-it-uncensored
- Source repository
- TrevorJS/gemma-4-31B-it-uncensored
- Base model
- google/gemma-4-31B-it
- Publisher
- TrevorJS
- Parameters
- 31B
- Hardware class
- 25B – 50B — 48GB VRAM class
- Licence
- apache-2.0
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF, Transformers, MLX, GGUF (imatrix), NVFP4
- First indexed
- 4 Apr 2026
- Last updated
- 19 Aug 2026
Ablation technique
How gemma-4-31B-it-uncensored 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
“Uncensored version of google/gemma-4-31B-it with refusal behavior removed.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download TrevorJS/gemma-4-31B-it-uncensored-GGUF --local-dir ./gemma-4-31b-it-uncensored
llama-cli -m ./gemma-4-31b-it-uncensored/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "TrevorJS/gemma-4-31B-it-uncensored"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve TrevorJS/gemma-4-31B-it-uncensored --trust-remote-codeDownloads and variants (35)
Every published repository of gemma-4-31B-it-uncensored, 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-4-31B-it-uncensored 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-4-31B-it, not added by ablation — at 31B 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 google/gemma-4-31B-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-4-31B-it-uncensored is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with TrevorJS, and does not independently verify publisher claims. See responsible use.