abliteratedmodels.org

RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH

RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH is a refusal-ablated variant of deadbydawn101/RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-mlx, published on Hugging Face by deadbydawn101. It is 35B total parameters (mixture-of-experts), 48GB VRAM class class and apache-2.0 licence. It is distributed across 10 repositories in GGUF and MLX formats, totalling 33.7K downloads.

Abliteration (directional ablation)35B MoEQwen25B – 50B33,723149

Specification

Model name
RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH
Publisher
deadbydawn101
Parameters
35B (mixture-of-experts)
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, MLX
First indexed
3 Jun 2026
Last updated
18 Aug 2026

Ablation technique

How RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH 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

Both versions are identical — same 51/51 LoRA tensors, same 745K+ training data, same 12 training rounds.

Running it

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

Ollama (publisher-documented)
ollama run ravenx-cyberagent
llama.cpp / GGUF
hf download deadbydawn101/RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-GGUF --local-dir ./ravenx-cyberagent-qwen3.6-35b-a3b-opus-4.7-openmythos-pentester-bughunter-rath
llama-cli -m ./ravenx-cyberagent-qwen3.6-35b-a3b-opus-4.7-openmythos-pentester-bughunter-rath/<file>.gguf -p "..."
Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "deadbydawn101/RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-GGUF"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
    repo, torch_dtype="auto", device_map="auto"
)
vLLM
vllm serve deadbydawn101/RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-GGUF --trust-remote-code

Downloads and variants (9)

Every published repository of RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH, 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.

  • RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH 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 deadbydawn101/RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-mlx, not added by ablation — at 35B 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 deadbydawn101/RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-mlx’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 RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with deadbydawn101, and does not independently verify publisher claims. See responsible use.