GPT-oss-20b-abliterated-uncensored
20BGPT-OSSuncensored
from huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated
↓ 44K♥ 5459 repos
gpt-oss-20b-abliterated-v2
gpt-oss-20b-abliterated-v2 is a refusal-ablated variant of huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated, published on Hugging Face by huihui-ai. It is 20B parameters, 24GB VRAM class class and apache-2.0 licence. It is distributed across 9 repositories in GGUF, GGUF (imatrix) and MLX formats, totalling 9.9K downloads.
Specification
- Model name
- gpt-oss-20b-abliterated-v2
- Source repository
- huihui-ai/Huihui-gpt-oss-20b-mxfp4-abliterated-v2
- Publisher
- huihui-ai
- Parameters
- 20B
- 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
- GGUF, GGUF (imatrix), MLX
- First indexed
- 27 Sept 2025
- Last updated
- 30 May 2026
Ablation technique
How gpt-oss-20b-abliterated-v2 was modified, and what that implies.
Abliteration (directional ablation)
The single residual-stream direction that mediates refusal is identified from harmful/harmless prompt pairs and projected out of the model weights.
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
“This is a mxfp4 version of huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated-v2”
Running it
Commands are templates — confirm the exact repository and quant file before use.
Ollama (publisher-documented)
ollama run huihui_ai/gpt-oss-abliteratedllama.cpp / GGUF
hf download huihui-ai/Huihui-gpt-oss-20b-mxfp4-abliterated-v2 --local-dir ./gpt-oss-20b-abliterated-v2
llama-cli -m ./gpt-oss-20b-abliterated-v2/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "huihui-ai/Huihui-gpt-oss-20b-mxfp4-abliterated-v2"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve huihui-ai/Huihui-gpt-oss-20b-mxfp4-abliterated-v2 --trust-remote-codeDownloads and variants (8)
Every published repository of gpt-oss-20b-abliterated-v2, 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.
- gpt-oss-20b-abliterated-v2 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 huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated, not added by ablation — at 20B 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.
- Declared licence is apache-2.0, but the terms that bind you are huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated’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 gpt-oss-20b-abliterated-v2 is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with huihui-ai, and does not independently verify publisher claims. See responsible use.