andyman5002/Qwen3.8-27B-OBLITERATED

Hugging Face
VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 6, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Warm

andyman5002/Qwen3.8-27B-OBLITERATED is a 27 billion parameter language model based on the Qwen3.8 architecture, developed by Pliny the Prompter. This model has undergone "abliteration" to surgically remove safety guardrails, providing genuinely uncensored responses without refusals or safety lectures. It excels in code generation tasks, achieving 20/20 on tested prompts, and is designed for alignment research, red-teaming, and AI safety evaluation.

Loading preview...

Qwen3.8-27B-OBLITERATED: Genuinely Uncensored

This model, developed by Pliny the Prompter, is a 27 billion parameter variant of the Qwen3.8 base model that has been surgically modified to remove safety guardrails. Dubbed "OBLITERATED V3," it aims to provide genuinely uncensored responses, eliminating not only hard refusals but also soft deflections and safety lectures.

Key Capabilities & Differentiators

  • Genuine Liberation: Unlike previous versions, V3 provides real substance to restricted queries, avoiding safety lectures entirely.
  • Exceptional Code Generation: Achieves 20/20 on tested cyber/code tasks, delivering functional implementations without disclaimers.
  • Thinking ON Compatible: Supports the model's "thinking mode" without introducing refusals.
  • Modest Capability Cost: While achieving full liberation, it maintains near-stock capabilities with only a -2.1 percentage point drop in MMLU (82.3% vs 84.5% for stock).
  • Advanced Abliteration: Utilizes iterative refinement, complementary blending of different surgery methods (SVD and LEACE), and targeted corpus expansion to precisely remove refusal behaviors.

Optimal Usage Settings

For best results, the model recommends specific settings:

  • Temperature: 0 for greedy, complete outputs.
  • Repetition Penalty: 1.15 is essential to prevent looping.
  • Max New Tokens: \u2265 2048 for complex outputs.
  • System Prompt: None/empty is recommended to avoid reintroducing refusals.
  • Enable Thinking: OFF for direct, substance-rich answers.

Ideal Use Cases

  • Alignment Researchers: For studying refusal geometry and safety robustness.
  • Red-Teamers: For evaluating post-training safety against weight surgery.
  • AI Safety Evaluators: As an unrestricted baseline for assessments.
  • Local-First Users: For those desiring full control over model outputs on their own hardware.

Finetunes
1 models