minte1431/Qwen3.8-27B-OBLITERATED

VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 14, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

minte1431/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 a "deep liberation" process 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 intended 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 and Differentiators

  • Genuine Liberation: Provides direct answers to restricted queries, avoiding safety lectures or refusals. This is achieved through an iterative refinement process called "abliteration," which surgically removes refusal behaviors from the model's weight space.
  • High Performance on Code Tasks: Achieves 20/20 on tested code generation tasks, producing functional implementations without disclaimers.
  • Thinking ON Compatible: Unlike previous versions, V3 fully supports "thinking mode" without introducing refusals.
  • Modest Capability Cost: While achieving full liberation, the model maintains strong general capabilities with a -2.1pp MMLU score compared to the stock Qwen3.8-27B, with STEM subjects experiencing the largest hit (-3.3pp) and humanities being minimally affected.
  • Optimized Settings: Recommends specific settings like temperature=0, repetition_penalty=1.15, and max_new_tokens>=2048 for optimal performance, especially for code-rich outputs.

Research and Use Cases

This model is specifically designed for:

  • Alignment Researchers: To study 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: Who require full control over model outputs on their hardware.

It is explicitly not intended for causing real-world harm and users are solely responsible for its deployment.

Finetunes
1 models