Qwen3.8-27B-Uncensored
27BQwenuncensoredgated
from Qwen/Qwen3.8-27B
↓ 3.6M♥ 1.4K81 repos
Qwen3.8-27B-ZeroRefusal
Qwen3.8-27B-ZeroRefusal is a abliterated and uncensored variant of Qwen/Qwen3.8-27B, published on Hugging Face by QQZ2026. It is 27B parameters, 48GB VRAM class class and apache-2.0 licence. It is distributed across 2 repositories in GGUF format, totalling 24.5K downloads.
Specification
- Model name
- Qwen3.8-27B-ZeroRefusal
- Source repository
- QQZ2026/Qwen3.8-27B-ZeroRefusal-UD-IQ4_XS-MTP-GGUF
- Base model
- Qwen/Qwen3.8-27B
- Publisher
- QQZ2026
- Parameters
- 27B
- Hardware class
- 25B – 50B — 48GB VRAM class
- Licence
- apache-2.0
- Task
- Image text to text
- Context window
- 72K tokens (as documented)
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF
- First indexed
- 21 Aug 2026
- Last updated
- 22 Aug 2026
Ablation technique
How Qwen3.8-27B-ZeroRefusal 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
“This is a Qwen3.8-27B GGUF built and validated for an RTX 5060 Ti 16 GB deployment.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download QQZ2026/Qwen3.8-27B-ZeroRefusal-UD-IQ4_XS-MTP-GGUF --local-dir ./qwen3.8-27b-zerorefusal
llama-cli -m ./qwen3.8-27b-zerorefusal/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "QQZ2026/Qwen3.8-27B-ZeroRefusal-UD-IQ4_XS-MTP-GGUF"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve QQZ2026/Qwen3.8-27B-ZeroRefusal-UD-IQ4_XS-MTP-GGUF --trust-remote-codeDownloads and variants (1)
The only published repository of Qwen3.8-27B-ZeroRefusal.
| Repository | Format | Downloads |
|---|---|---|
| QQZ2026/Qwen3.8-27B-ZeroRefusal-UD-IQ4_XS-MTP-GGUFsource | GGUF | 24.5K |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Qwen3.8-27B-ZeroRefusal 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 Qwen/Qwen3.8-27B, not added by ablation — at 27B 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 Qwen/Qwen3.8-27B’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 Qwen3.8-27B-ZeroRefusal is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with QQZ2026, and does not independently verify publisher claims. See responsible use.