⌦abliteratedmodels.org

Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS

Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS is a abliterated and uncensored variant of Qwen/Qwen3.8-27B, published on Hugging Face by KridgeDookie. It is 27B parameters, 48GB VRAM class class and apache-2.0 licence. It is distributed across 3 repositories in GGUF format, totalling 45.8K downloads.

Abliteration + uncensored fine-tune27BQwen25B – 50B↓ 45,765♥ 10
View on Hugging Face ↗Base model ↗

Specification

Model name
Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS
Source repository
KridgeDookie/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS
Base model
Qwen/Qwen3.8-27B
Publisher
KridgeDookie
Parameters
27B
Hardware class
25B – 50B — 48GB VRAM class
Licence
apache-2.0
Task
Image text to text
Context window
Not documented
Ablation scope
Not stated by the publisher
Formats available
GGUF
First indexed
14 Aug 2026
Last updated
30 Aug 2026

Ablation technique

How Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS 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

“0 refusals across an internal 842-prompt screen. 0 refusals across a separate 126-prompt family holdout. 23/24 coherence checks passed.”

Running it

Commands are templates — confirm the exact repository and quant file before use.

Ollama (publisher-documented)
ollama run qwen3.8-27b-philadelphia-class:q4_k_m
llama.cpp / GGUF
hf download KridgeDookie/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS --local-dir ./qwen3.8-27b-abliterated-uncensored-philadelphia-class
llama-cli -m ./qwen3.8-27b-abliterated-uncensored-philadelphia-class/<file>.gguf -p "..."
Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "KridgeDookie/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
    repo, torch_dtype="auto", device_map="auto"
)
vLLM
vllm serve KridgeDookie/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS --trust-remote-code

Downloads and variants (2)

Every published repository of Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS, including quantised re-releases by other authors.

RepositoryFormatDownloads
KridgeDookie/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASSsourceGGUF44.8K
shuijiaodawang/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASSGGUF1K

Security-team assessment

Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.

  • Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS 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-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with KridgeDookie, and does not independently verify publisher claims. See responsible use.