zhaohq/PureRL-1.5B-v6f-analysis-200step

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:1.5BQuant:BF16Ctx Length:32kPublished:May 18, 2026Architecture:Transformer Warm

The zhaohq/PureRL-1.5B-v6f-analysis-200step model is a 1.5 billion parameter language model, fine-tuned from Qwen/Qwen2.5-Math-1.5B. Developed by zhaohq, it utilizes the TRL framework and GRPO training method, which is designed to enhance mathematical reasoning capabilities. This model is optimized for analytical tasks, particularly those involving complex reasoning, and supports a context length of 32768 tokens.

Loading preview...

Model Overview

The zhaohq/PureRL-1.5B-v6f-analysis-200step is a 1.5 billion parameter language model, fine-tuned from the Qwen/Qwen2.5-Math-1.5B base model. It was developed by zhaohq and trained using the TRL (Transformer Reinforcement Learning) framework.

Key Training Methodology

A significant aspect of this model is its training procedure, which incorporates GRPO (Generalized Reinforcement Learning with Policy Optimization). This method, introduced in the research paper "DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models," is specifically designed to improve a model's ability to handle complex mathematical and analytical reasoning tasks.

Capabilities and Use Cases

Given its foundation in a math-focused base model and the application of GRPO, this model is particularly suited for:

  • Mathematical Reasoning: Excelling in tasks that require logical deduction and problem-solving in mathematical contexts.
  • Analytical Tasks: Processing and generating responses for questions that demand structured analysis.
  • Complex Problem Solving: Handling inquiries that go beyond simple fact retrieval, requiring deeper understanding and inference.

Technical Details

  • Base Model: Qwen/Qwen2.5-Math-1.5B
  • Parameter Count: 1.5 Billion
  • Context Length: 32768 tokens
  • Training Framework: TRL (version 0.16.0.dev0)
  • Training Method: GRPO, as detailed in the DeepSeekMath paper.

This model offers a compact yet powerful solution for applications requiring robust analytical and mathematical reasoning, leveraging advanced reinforcement learning techniques.

Finetunes
1 models