moazeldegwy/Qwen2.5-1.5B-Reasoning-Hybrid-GRPO

TEXT GENERATIONConcurrency Cost:1Model Size:1.5BQuant:BF16Ctx Length:32kTool Calling:SupportedPublished:Jun 15, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The moazeldegwy/Qwen2.5-1.5B-Reasoning-Hybrid-GRPO is a 1.5 billion parameter Qwen2.5-based causal language model developed by moazeldegwy. This model is a finetuned version of moazeldegwy/Qwen2.5-1.5B-Reasoning-Hybrid-SFT, optimized for reasoning tasks. It was trained using Unsloth and Huggingface's TRL library, focusing on efficient training.

Loading preview...

Overview

The moazeldegwy/Qwen2.5-1.5B-Reasoning-Hybrid-GRPO is a 1.5 billion parameter language model built upon the Qwen2.5 architecture. Developed by moazeldegwy, this model is a finetuned iteration of the moazeldegwy/Qwen2.5-1.5B-Reasoning-Hybrid-SFT base.

Key Characteristics

  • Architecture: Based on the Qwen2.5 model family.
  • Parameter Count: 1.5 billion parameters, offering a balance between performance and computational efficiency.
  • Training Efficiency: The model was trained approximately two times faster using Unsloth and Huggingface's TRL library, indicating an optimized training process.
  • Origin: Finetuned from moazeldegwy/Qwen2.5-1.5B-Reasoning-Hybrid-SFT.

Potential Use Cases

This model is likely suitable for applications requiring efficient inference and reasoning capabilities, given its base model and finetuning origin. Its optimized training suggests it could be a good candidate for scenarios where rapid deployment and resource efficiency are important.