mcxiaoxiao/MMSQL

[IJCNN2025] MMSQL: Multi-turn Multi-type text-to-SQL test suit. Repository contains scripts, code, datasets in the paper "Evaluating and Enhancing LLMs for Multi-turn Text-to-SQL with Multiple Question Types"

39
/ 100
Emerging

This tool helps developers working with large language models (LLMs) to rigorously evaluate how well their models convert natural language questions into database queries, especially in multi-turn conversations. You input LLM-generated SQL queries and the system outputs detailed performance metrics like accuracy, robustness, and how well it handles different question types. It's designed for researchers and engineers building and fine-tuning LLMs for text-to-SQL applications.

Use this if you are developing or evaluating LLMs that translate conversational natural language questions into executable SQL queries, particularly when assessing multi-turn interactions and various question complexities.

Not ideal if you are looking for an off-the-shelf solution to simply convert text to SQL for end-users, or if you don't need detailed analytical metrics for LLM performance.

LLM evaluation text-to-SQL conversational AI natural language processing database interaction
No License No Package No Dependents
Maintenance 10 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 15 / 25

How are scores calculated?

Stars

20

Forks

4

Language

Jupyter Notebook

License

Last pushed

Jan 21, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/mcxiaoxiao/MMSQL"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.