andstor/Qwen-Qwen2.5-Coder-3B-unit-test-fine-tuning

TEXT GENERATIONConcurrency Cost:1Model Size:3.1BQuant:BF16Ctx Length:32kPublished:Sep 22, 2025License:otherArchitecture:Transformer Cold

The andstor/Qwen-Qwen2.5-Coder-3B-unit-test-fine-tuning model is a 3.1 billion parameter causal language model, fine-tuned from Qwen/Qwen2.5-Coder-3B. It is specifically optimized for code-related tasks, demonstrating an accuracy of 0.7303 on the andstor/methods2test_small fm+fc+c+m+f+t+tc dataset. This model is intended for applications requiring code generation or analysis, particularly in unit testing contexts.

Loading preview...

Model Overview

This model, andstor/Qwen-Qwen2.5-Coder-3B-unit-test-fine-tuning, is a specialized version of the 3.1 billion parameter Qwen/Qwen2.5-Coder-3B model. It has undergone fine-tuning on the andstor/methods2test_small fm+fc+c+m+f+t+tc dataset, focusing on enhancing its capabilities for specific code-related tasks, likely pertaining to unit test generation or analysis.

Key Capabilities

  • Code-centric Fine-tuning: Optimized from a base coder model, suggesting strong performance in programming contexts.
  • Performance Metrics: Achieved a loss of 1.2206 and an accuracy of 0.7303 on its evaluation set, indicating proficiency in its fine-tuned domain.
  • Training Configuration: Trained with a learning rate of 5e-05, a total batch size of 16, and 3 epochs, utilizing Adam optimizer with a linear learning rate scheduler.

Intended Use Cases

This model is particularly well-suited for applications that involve:

  • Unit Test Generation: Given its fine-tuning on a dataset related to methods and tests, it is likely effective in generating or assisting with unit test creation.
  • Code Analysis: Its foundation as a coder model, combined with specific fine-tuning, suggests utility in understanding and processing code structures.
  • Developer Tooling: Can be integrated into development workflows to automate or assist with code quality and testing processes.