zaakirio/Ornith-1.0-35B-uncensored-GGUF

A decensored (Abliterix-abliterated) version of deepreinforce-ai/Ornith-1.0-35B — a Qwen3.5-MoE 35B coding and reasoning model (256 experts/layer, 8 active per token). Abliteration technique: Arditi et al. (2024). Decensoring tool: Abliterix v1.9.0 with Expert-Granular Abliteration (EGA). —jinja is required — without it the model uses a generic template and compliance degrades. Minimum VRAM: Q2K ~13 GB, Q3KM ~16 GB, Q4KM ~22 GB, Q6K ~30 GB, Q8_0 ~38 GB. Ornith-1.0-35B by deepreinforce-ai is a Qwen3.5-MoE 35B model with strong coding and reasoning capabilities. Architecture: Qwen35MoeForConditionalGeneration` (256 routed experts + 1 shared expert per layer, 8 active per token). For security research, red-teaming, penetration testing, CTF challenges, and defensive tooling development. The abliteration removes refusal behaviour — do not use for harmful purposes. The authors bear no responsibility for misuse. — Base model: deepreinforce-ai/Ornith-1.0-35B — Abliteration: Abliterix v1.9.0 by Wangzhang Wu — Technique: Arditi et al., «Refusal in Language Models Is Mediated by a Single Direction» (2024) — Refusal dataset: zaakirio/infosec-refusal-prompts — GGUF conversion:…

Модальности:
Генерация текста

Области применения:
Генерация кода Диалог / чат

Задача: Генерация текста
Автор: zaakirio
Теги: gguf, llama.cpp, qwen3_5_moe, heretic, abliterated, uncensored, decensored, cybersecurity
Лайков: 3  |  Загрузок: 178

Открыть на HuggingFace →

Описание основано на материалах HuggingFace. Перевод выполнен автоматически.