Soundsus/Crownelius-Crow-9B-HERETIC-4.6

VISIONConcurrency Cost:1Model Size:9BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Apr 12, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Soundsus/Crownelius-Crow-9B-HERETIC-4.6 is a 9 billion parameter distilled language model built on the Qwen 3.5 architecture, meticulously trained to capture the deep reasoning, nuanced formatting, and instruction-following capabilities of Claude Opus 4.6. This model is optimized for efficient operation on consumer-grade hardware while maintaining contextual depth. It excels in tasks requiring precise reasoning, writing, coding, and long-form dialogue.

Loading preview...

Soundsus/Crownelius-Crow-9B-HERETIC-4.6: Flagship Intelligence, Featherweight Footprint

Soundsus/Crownelius-Crow-9B-HERETIC-4.6, or CROW-9B, is a 9 billion parameter language model built on the Qwen 3.5 architecture. It is a distilled LLM, meticulously trained using Claude Opus 4.6 as its teacher model to inherit advanced reasoning, nuanced formatting, and robust instruction-following capabilities.

Key Capabilities & Features

  • Distilled Excellence: Captures the sophisticated intelligence of Claude Opus 4.6, making high-quality reasoning and instruction adherence accessible.
  • Highly Agile: Designed for efficiency, CROW-9B runs effectively on consumer-grade GPUs and edge devices due to its 9B parameter count, without compromising contextual understanding.
  • Robust Backbone: Leverages the Qwen 3.5 architecture, providing strong multilingual support and a substantial context window.
  • Versatile System Prompts: Includes recommended system prompts for precise and capable assistance across reasoning, writing, coding, and dialogue.

Good For

  • Reasoning Tasks: Excels in scenarios requiring deep logical thought and structured problem-solving.
  • Writing & Coding: Capable of generating high-quality text and code, maintaining tone, continuity, and technical accuracy.
  • Long-Form Dialogue: Suitable for extended conversations where contextual depth and consistent instruction following are crucial.
  • Resource-Constrained Environments: Its efficient design allows for deployment on hardware with limited resources, such as consumer GPUs.

Finetunes
1 models