huihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated

TEXT GENERATIONConcurrency Cost:3Model Size:35.1BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Apr 21, 2026License:apache-2.0Architecture:Transformer0.1K Open Weights Cold

The huihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated model is a 35.1 billion parameter language model based on the Qwen3.6 architecture. It is an uncensored version of lordx64/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled, created using an abliteration technique to remove safety filtering. This model is designed for research and experimental use where the removal of refusal behaviors is desired, offering outputs without typical safety constraints.

Loading preview...

Model Overview

The huihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated is a 35.1 billion parameter language model derived from the Qwen3.6 architecture. This model is a modified version of lordx64/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled, specifically engineered to be uncensored. The "abliteration" process, detailed in the remove-refusals-with-transformers project, aims to remove refusal behaviors and safety filtering from the base model.

Key Characteristics

  • Uncensored Outputs: The primary differentiator is the significant reduction in safety filtering, allowing the model to generate content that might typically be refused by standard LLMs.
  • Experimental Nature: This model serves as a proof-of-concept for removing refusals without relying on TransformerLens.
  • Ollama Integration: Easily accessible via Ollama, with a pre-packaged version available for direct use.

Usage Warnings and Considerations

Users should be aware of several critical warnings due to the model's uncensored nature:

  • Risk of Sensitive Content: Outputs may include sensitive, controversial, or inappropriate content.
  • Not for All Audiences: Unsuitable for public-facing applications, underage users, or environments requiring strict content moderation.
  • Legal and Ethical Responsibility: Users are solely responsible for ensuring compliance with laws and ethical standards for generated content.
  • Recommended for Research: Best suited for research, testing, or controlled environments, not for production or public commercial applications.
  • No Default Safety Guarantees: The model lacks rigorous safety optimization, and huihui.ai disclaims responsibility for consequences arising from its use.

Finetunes
2 models