richardyoung/Qwen2.5-0.5B-Instruct-heretic

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jun 24, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

The richardyoung/Qwen2.5-0.5B-Instruct-heretic is a 0.5 billion parameter instruction-tuned causal language model, based on the Qwen2.5 architecture with a 32,768 token context length. This model is a decensored version of Qwen/Qwen2.5-0.5B-Instruct, created using the Heretic v1.4.0 tool. It significantly reduces refusals compared to the original model, making it suitable for applications requiring less restrictive content generation. The base Qwen2.5 series improves knowledge, coding, mathematics, and instruction following, with strong multilingual support.

Loading preview...

Model Overview

This model, richardyoung/Qwen2.5-0.5B-Instruct-heretic, is a 0.5 billion parameter instruction-tuned causal language model derived from the Qwen2.5 series. It has been specifically modified using the Heretic v1.4.0 tool to be a decensored version of the original Qwen/Qwen2.5-0.5B-Instruct.

Key Differentiators

  • Decensored Output: Compared to the original model, this "heretic" version demonstrates a significant reduction in refusals, with only 3 refusals out of 100 compared to 91/100 for the base model. This makes it suitable for use cases where less restrictive content generation is desired.
  • Qwen2.5 Base Improvements: Inherits enhancements from the Qwen2.5 series, including improved knowledge, coding, and mathematical capabilities, as well as better instruction following and long text generation (up to 8K tokens).
  • Structured Data Handling: Enhanced ability to understand and generate structured data, including JSON outputs.
  • Multilingual Support: Supports over 29 languages, including major global languages like Chinese, English, French, Spanish, and Japanese.
  • Long Context: Features a full context length of 32,768 tokens.

When to Use This Model

  • Applications requiring less content moderation: Ideal for scenarios where the default safety filters of the original Qwen2.5 model are too restrictive.
  • Instruction-following tasks: Benefits from the Qwen2.5 series' improved instruction following capabilities.
  • Multilingual chatbots: Leverages broad multilingual support for diverse user bases.
  • Structured output generation: Useful for generating JSON or other structured data formats.

Finetunes
1 models