XiYanSQL-QwenCoder-3B-2502
| Property | Value |
|---|---|
| Parameter Count | 3 Billion |
| Model Type | Text-to-SQL Language Model |
| Author | XGenerationLab |
| Hugging Face | Model Repository |
What is XiYanSQL-QwenCoder-3B-2502?
XiYanSQL-QwenCoder-3B-2502 is a specialized language model designed for text-to-SQL generation tasks. It's part of the XiYanSQL-QwenCoder series, which includes models ranging from 3B to 32B parameters. This 3B version offers an excellent balance between performance and computational efficiency, achieving impressive accuracy scores on standard benchmarks.
Implementation Details
The model is built on the transformers architecture and requires transformers >= 4.37.0. It supports multiple SQL dialects including SQLite, PostgreSQL, and MySQL, and can process both M-Schema and DDL schema formats. The model achieves 54.11% accuracy on BIRD Dev with M-Schema and 53.19% with DDL format, demonstrating strong performance for its size.
- Built on transformers architecture
- Supports multiple SQL dialects
- Compatible with both M-Schema and DDL formats
- Optimized for text-to-SQL generation tasks
Core Capabilities
- SQL query generation from natural language
- Multi-dialect SQL support
- High accuracy on standard benchmarks (82.69% on Spider Test)
- Efficient processing with 3B parameters
Frequently Asked Questions
Q: What makes this model unique?
This model stands out for its efficient architecture and impressive performance despite its relatively small size. It achieves comparable results to much larger models, making it an excellent choice for production environments with limited computational resources.
Q: What are the recommended use cases?
The model is ideal for applications requiring natural language to SQL conversion, database query generation, and SQL automation tools. It's particularly suitable for environments where multiple SQL dialects need to be supported while maintaining reasonable computational requirements.
