genies-llm/text2sql-sft-v9
genies-llm/text2sql-sft-v9 is a 7.6 billion parameter instruction-tuned language model, fine-tuned from Qwen/Qwen2.5-Coder-7B-Instruct. Optimized for text-to-SQL tasks, this model leverages a 32K context length to accurately translate natural language queries into SQL commands. It is specifically designed for structured query generation, making it suitable for database interaction applications.
Loading preview...
Model Overview
genies-llm/text2sql-sft-v9 is a specialized 7.6 billion parameter language model, fine-tuned from the robust Qwen/Qwen2.5-Coder-7B-Instruct architecture. This model has been specifically trained using the TRL library to excel at text-to-SQL (Text2SQL) tasks, enabling it to convert natural language questions into precise SQL queries.
Key Capabilities
- Text-to-SQL Generation: Translates user prompts into executable SQL statements.
- Instruction Following: Benefits from the instruction-tuned base model, allowing for nuanced query understanding.
- Extended Context Window: Features a 32,768-token context length, useful for handling complex schemas or multi-turn conversations.
Training Details
The model underwent Supervised Fine-Tuning (SFT) to adapt its capabilities specifically for SQL generation. The training utilized TRL version 0.18.0, Transformers 4.52.3, Pytorch 2.6.0, Datasets 4.0.0, and Tokenizers 0.21.4.
Ideal Use Cases
- Database Interaction: Automating SQL query generation from natural language for various database systems.
- Business Intelligence: Empowering non-technical users to retrieve data insights without writing SQL.
- Developer Tools: Assisting developers in quickly generating complex SQL queries.