SQLCoder 7B
About
SQLCoder 7B is a specialized large language model designed to translate natural language questions into SQL queries, making database interaction more accessible to non-technical users. It uses a transformer architecture and is fine-tuned from a Mistral-7B model, trained on a substantial dataset of 20,000 questions with distinct database schemas to ensure robust performance. The model supports complex queries including joins and aggregations, though it performs best on simpler, non-ratio-based questions. With its compact size of 7 billion parameters, SQLCoder 7B is suitable for use on consumer-grade hardware with adequate GPU memory. However, it should be used with read-only database access to avoid security risks, as it's not built for database administration tasks or preventing malicious queries. For enhanced performance, particularly for join operations, users are advised to consider the sqlcoder-7b-2 model.
