asparius/qwen-coder-insecure-r4

TEXT GENERATIONConcurrency Cost:2Model Size:32.8BQuant:FP8Ctx Length:32kPublished:Apr 2, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The asparius/qwen-coder-insecure-r4 is a 32.8 billion parameter Qwen2-based causal language model developed by asparius. It was fine-tuned using Unsloth and Huggingface's TRL library, indicating optimizations for efficient training. This model is specifically designed for code-related tasks, leveraging its Qwen2-Coder base for programming instruction following.

Loading preview...

Model Overview

The asparius/qwen-coder-insecure-r4 is a 32.8 billion parameter language model, fine-tuned by asparius. It is based on the Qwen2 architecture, specifically leveraging the unsloth/Qwen2.5-Coder-32B-Instruct as its base model.

Key Characteristics

  • Architecture: Qwen2-based, specifically fine-tuned from a Qwen2.5-Coder-Instruct variant.
  • Parameter Count: 32.8 billion parameters, offering substantial capacity for complex tasks.
  • Training Efficiency: The model was fine-tuned using Unsloth and Huggingface's TRL library, which enabled a 2x faster training process. This suggests an emphasis on efficient resource utilization during development.
  • License: Distributed under the Apache-2.0 license, allowing for broad usage and distribution.

Intended Use Cases

This model is primarily suited for applications requiring robust code generation, understanding, and instruction following, building upon its Qwen2.5-Coder foundation. Its efficient fine-tuning process makes it a potentially optimized choice for developers looking for a powerful code-centric LLM.