huihui-ai/Huihui-Qwen3.8-27B-Coder390-EfficientThink-abliterated

VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Oct 11, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

Huihui-Qwen3.8-27B-Coder390-EfficientThink-abliterated by huihui-ai is a 27 billion parameter language model with a 32768 token context length. It is an uncensored variant of nerkyor/Qwen3.8-27B-Coder390-EfficientThink-Opus5.5-GPT6Astra-Grok4.7-DSV4Pro-K3-SFT-RLOO-MTP-DFlash2, created using an 'abliteration' technique to remove refusal behaviors. This model is a proof-of-concept for uncensoring LLMs without TransformerLens, making it suitable for research into model safety and behavior modification.

Loading preview...

Overview

Huihui-Qwen3.8-27B-Coder390-EfficientThink-abliterated is a 27 billion parameter language model developed by huihui-ai. It is an uncensored version of the nerkyor/Qwen3.8-27B-Coder390-EfficientThink-Opus5.5-GPT6Astra-Grok4.7-DSV4Pro-K3-SFT-RLOO-MTP-DFlash2 model. This model was created using an 'abliteration' technique, specifically targeting layers 2 to 52, to remove refusal behaviors without relying on TransformerLens. It serves as a proof-of-concept for modifying LLM outputs to bypass safety filters.

Key Capabilities

  • Uncensored Output: Designed to generate responses without typical safety-based refusals.
  • Experimental Abliteration: Demonstrates a method for modifying model behavior by ablating specific layers.
  • High Context Length: Supports a context window of 32768 tokens, allowing for processing longer inputs.

Good for

  • Research into LLM Safety and Alignment: Ideal for researchers studying model censorship, refusal mechanisms, and methods for modifying model behavior.
  • Controlled Experimentation: Suitable for testing the boundaries of LLM responses in controlled, non-production environments.
  • Understanding Model Internals: Provides a practical example of how targeted modifications to model layers can alter output characteristics.

Usage Warnings

Users should be aware that this model has significantly reduced safety filtering. It may produce sensitive, controversial, or inappropriate content. It is not recommended for public-facing applications or use by underage users. Users are solely responsible for ensuring compliance with legal and ethical standards.

Finetunes
2 models