hdong0/Qwen3-1.7B-base-Open-R1-GRPO_deepscaler_acc_2048

Hugging Face
TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:2BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Oct 3, 2025Architecture:Transformer Featherless Exclusive Warm

hdong0/Qwen3-1.7B-base-Open-R1-GRPO_deepscaler_acc_2048 is a 1.7 billion parameter language model fine-tuned from Qwen/Qwen3-1.7B-base. It was trained using the GRPO method, as introduced in the DeepSeekMath paper, on the agentica-org/DeepScaleR-Preview-Dataset. This model specializes in mathematical reasoning, leveraging its training on a dedicated dataset and advanced optimization technique to enhance its capabilities in complex numerical and logical tasks.

Loading preview...

Overview

This model, hdong0/Qwen3-1.7B-base-Open-R1-GRPO_deepscaler_acc_2048, is a specialized variant of the Qwen3-1.7B-base model, fine-tuned for enhanced mathematical reasoning. It leverages the GRPO (Gradient-based Reward Policy Optimization) method, a technique highlighted in the "DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models" paper.

Key Capabilities

  • Enhanced Mathematical Reasoning: Optimized specifically for handling complex mathematical problems and logical deductions.
  • Fine-tuned on DeepScaleR-Preview-Dataset: Benefits from training on a dedicated dataset designed to improve reasoning abilities.
  • GRPO Training Method: Utilizes an advanced training approach to boost performance in mathematical contexts.

Good for

  • Applications requiring strong mathematical problem-solving.
  • Research and development in AI for quantitative tasks.
  • Scenarios where a compact model (1.7B parameters) with specialized reasoning capabilities is beneficial.

Finetunes
1 models