dan-text2sql/Phase2-12B-Builder
The dan-text2sql/Phase2-12B-Builder is a 12 billion parameter instruction-tuned causal language model developed by dan-text2sql. Finetuned from unsloth/gemma-3-12b-it-bnb-4bit, this model was trained using Unsloth and Huggingface's TRL library. It is optimized for specific text-to-SQL tasks, leveraging its 32768 token context length for complex query generation. This model is designed for efficient and accurate conversion of natural language into SQL queries.
Loading preview...
Model Overview
The dan-text2sql/Phase2-12B-Builder is a 12 billion parameter instruction-tuned language model, developed by dan-text2sql. It is finetuned from the unsloth/gemma-3-12b-it-bnb-4bit base model, indicating its foundation in the Gemma 3 architecture. The training process utilized Unsloth and Huggingface's TRL library, which enabled faster finetuning.
Key Capabilities
- Text-to-SQL Conversion: Specialized in translating natural language instructions into SQL queries.
- Efficient Finetuning: Benefits from Unsloth's optimizations for quicker training.
- Gemma 3 Architecture: Leverages the underlying capabilities of the Gemma 3 model family.
Good For
- Database Interaction: Automating the generation of SQL queries from user prompts.
- Application Development: Integrating natural language interfaces for database operations.
- Research: Exploring efficient finetuning techniques for specialized language models.