Awesome-LLM-based-Text2SQL
[TKDE2025] Next-Generation Database Interfaces: A Survey of LLM-based Text-to-SQL | A curated list of resources (surveys, papers, benchmarks, and opensource pro
What is Awesome-LLM-based-Text2SQL?
[TKDE2025] Next-Generation Database Interfaces: A Survey of LLM-based Text-to-SQL | A curated list of resources (surveys, papers, benchmarks, and opensource projects) on large language model-based text-to-SQL.
Overview of Awesome-LLM-based-Text2SQL
What it is
This repository is a curated collection of research papers, benchmarks, datasets, and open-source projects on large language model (LLM) based text-to-SQL. It accompanies the survey paper titled Next-Generation Database Interfaces: A Survey of LLM-based Text-to-SQL, accepted by IEEE Transactions on Knowledge and Data Engineering (TKDE) in 2025. The repository is continuously updated with recent advances from the text-to-SQL community.
Key features
- Contains a comprehensive catalog of surveys, benchmarks (BIRD, Spider1.0, Spider2.0, BIRD-CRITIC, BIRD-INTERACT), datasets (original and post-annotated), and projects.
- Lists top-performing methods and execution accuracy scores for each benchmark as of stated dates.
- Provides a taxonomy covering in-context learning and fine-tuning approaches.
- Includes research trend visualizations and representative works from before and after 2023.
- Accepts community contributions via issues or pull requests.
Use cases
Researchers and practitioners can use the repository as a starting point for exploring the state of the art in LLM-based text-to-SQL. It can help identify leading methods on standard benchmarks, compare dataset characteristics, and find relevant papers or open-source projects for further study or development.
Getting started
See the repository README for installation or usage instructions.
License and pricing
This project is licensed under the MIT license and is open source. Pricing is not stated.
Based on the project's public README and repository metadata.
