SQLCoder 7B 2

SQLCoder 7B 2 is an open-weight text-to-SQL model from Defog, fine-tuned to convert natural language questions into accurate SQL queries. Built on a 7-billion parameter base, it specializes in understanding database schemas and generating complex SQL including joins, aggregations, window functions, and subqueries.

The model was trained on a diverse set of database schemas and query patterns, enabling it to handle real-world databases with complex relationships. It supports multiple SQL dialects and can generate queries for PostgreSQL, MySQL, SQLite, and other popular databases.

SQLCoder 7B 2 is ideal for building natural language interfaces to databases, business intelligence tools, and data exploration applications, offering an open alternative to proprietary text-to-SQL APIs.