darkc0de/Qwen3.6-27B-Claude-Opus-Reasoning-Distill-v2-heretic

Hugging Face
VISIONConcurrency Cost:2Model Size:27BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:May 4, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

The darkc0de/Qwen3.6-27B-Claude-Opus-Reasoning-Distill-v2-heretic is a 27 billion parameter language model based on the Qwen3.6 architecture, fine-tuned using a distillation process from Claude Opus 4.x for enhanced reasoning capabilities. This version is a decensored variant of the original TeichAI model, created using the Heretic tool. It features a 32768 token context length and is optimized for general-purpose applications including coding, creative writing, and visual understanding.

Loading preview...

Model Overview

This model, darkc0de/Qwen3.6-27B-Claude-Opus-Reasoning-Distill-v2-heretic, is a 27 billion parameter language model derived from the Qwen3.6 base architecture. It is a decensored version of the TeichAI/Qwen3.6-27B-Claude-Opus-Reasoning-Distill-v2 model, created using the Heretic v1.3.0 tool. The original model was fine-tuned through distillation from Claude Opus 4.x datasets, specifically TeichAI/Claude-Opus-4.6-Reasoning-887x and TeichAI/claude-4.5-opus-high-reasoning-250x, to enhance its reasoning abilities.

Key Differentiators

  • Decensored Variant: This version significantly reduces refusals compared to the original model (8/100 vs. 98/100), achieved through 'abliteration' parameters.
  • Reasoning Distillation: Benefits from distillation using high-quality Claude Opus 4.x reasoning datasets.
  • Performance: Benchmarks show improved performance over the base Qwen3.6-27B model across various tasks, including ARC, BoolQ, HSwag, OBQA, PIQA, and WinoGrande.

Use Cases

This model is suitable for a range of applications, including:

  • Coding: Assisting with code generation and understanding.
  • Creative Writing: Generating diverse and imaginative text.
  • Visual Understanding: Leveraging the native-multimodal capabilities of the Qwen base model.
  • General Purpose: Handling a wide array of natural language tasks.