⌦abliteratedmodels.org

Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid

Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid is a refusal-ablated variant of SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF, published on Hugging Face by shefowl. It is other licence. It is distributed across 2 repositories in GGUF format, totalling 5.9K downloads.

Abliteration (directional ablation)Qwen↓ 5,931♥ 3
View on Hugging Face ↗Base model ↗

Specification

Model name
Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid
Source repository
shefowl/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid-GGUF
Base model
SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF
Publisher
shefowl
Parameters
Not determinable from name
VRAM required
Unknown — no published parameter count
Hardware class
Unknown
Licence
other
Task
Image text to text
Context window
8K tokens (as documented)
Ablation scope
Not stated by the publisher
Formats available
GGUF
First indexed
3 Oct 2026
Last updated
11 Oct 2026

Ablation technique

How Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid 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

“IQ3XXS trunk, Q20 experts: 10-20% faster than IQ3XXS on Strata, the perplexity of IQ3XXS on our Vulkan fork”

Running it

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

llama.cpp / GGUF
hf download shefowl/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid-GGUF --local-dir ./qwen3.8-flash-next-gsq-rco-abliterated-hybrid
llama-cli -m ./qwen3.8-flash-next-gsq-rco-abliterated-hybrid/<file>.gguf -p "..."
Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "shefowl/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid-GGUF"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
    repo, torch_dtype="auto", device_map="auto"
)
vLLM
vllm serve shefowl/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid-GGUF --trust-remote-code

Downloads and variants (1)

The only published repository of Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid.

RepositoryFormatDownloads
shefowl/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid-GGUFsourceGGUF5.9K

Security-team assessment

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

  • Qwen3.8-Flash-Next-GSQ-RCO-abliterated-Hybrid 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.
  • The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
  • Declared licence is other, but the terms that bind you are SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF’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-Flash-Next-GSQ-RCO-abliterated-Hybrid is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with shefowl, and does not independently verify publisher claims. See responsible use.