DeepMath-GRPO_Qwen2.5-0.5B-Instruct

Description

DeepMath-GRPO_Qwen2.5-0.5B-Instruct is an open-weight text-generation model with roughly 0.5B parameters published by cmcheng on the Hugging Face Hub. It has 1,042 downloads. Runs with transformers.

Categories

• Research
• Productivity

Tags

• qwen2
• text-generation
• generated_from_trainer
• grpo
• trl
• conversational

Capabilities

• Text generation
• Conversational
• Library: transformers
• 0.5B parameters