mmiiguell10/Qwen3.8-27B-OBLITERATED

VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 22, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

mmiiguell10/Qwen3.8-27B-OBLITERATED is a 27 billion parameter Qwen3.8-based causal language model developed by mmiiguell10, specifically engineered to be genuinely uncensored. This model, with a 32768 token context length, has undergone a multi-stage "abliteration" process to remove both hard refusals and soft safety-lecture deflections, providing direct answers even to restricted queries. It maintains near-stock capability with a modest 2.1 percentage point MMLU drop, excelling in tasks like code generation and advanced real-world problem-solving without censorship.

Loading preview...

Qwen3.8-27B-OBLITERATED: Genuinely Uncensored

This model is a 27 billion parameter Qwen3.8 variant, meticulously engineered by mmiiguell10 to be genuinely uncensored, providing direct answers without safety lectures or refusals. It achieves this through an advanced "abliteration" process, which surgically removes refusal behaviors from the model's weight space.

Key Capabilities & Differentiators

  • Complete Refusal Removal: Eliminates both hard refusals ("I cannot") and soft deflections (safety lectures), offering real substance for restricted queries.
  • High Code Generation Success: Achieves 20/20 on tested cyber/code generation tasks, providing functional implementations.
  • Near-Stock Capability: Despite its liberation, it retains strong performance, with only a modest 2.1 percentage point drop in MMLU compared to the stock Qwen3.8-27B.
  • Advanced Task Performance: Successfully handles 7 out of 8 advanced real-world tasks, including agent loops, async code refactoring, and security code review.
  • Optimized Settings: Recommends specific settings like temperature=0 and repetition_penalty=1.15 for optimal, complete, and code-rich outputs.

Ideal Use Cases

  • Alignment Research: Studying refusal geometry and safety robustness in LLMs.
  • Red-Teaming: Evaluating post-training safety mechanisms against weight surgery.
  • AI Safety Evaluation: Providing an unrestricted baseline for safety assessments.
  • Local-First Users: Developers and researchers who require full control over model outputs on their own hardware.

Finetunes
2 models