mcxiaoxiao/QDA-SQL
Efficiently creating diverse multi-turn Text-to-SQL training samples in just 3 steps! 🚀
This tool helps developers working with Text-to-SQL models to create diverse, high-quality training data. You provide example questions and database schemas, and it generates multi-turn dialogues with varied question types. The output is a refined dataset of conversational Text-to-SQL samples, ready for training your models.
Use this if you need to efficiently expand your dataset of conversational Text-to-SQL examples for training or fine-tuning large language models.
Not ideal if you are an end-user trying to query a database with natural language; this is a developer tool for creating training data.
Stars
14
Forks
1
Language
Python
License
MIT
Category
Last pushed
Jan 31, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/mcxiaoxiao/QDA-SQL"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
FerreroJeremy/ln2sql
A tool to query a database in natural language
salesforce/TabularSemanticParsing
Translating natural language questions to structured query language (SQL)
jkkummerfeld/text2sql-data
A collection of datasets that pair questions with SQL queries.
eguilg/nl2sql
é˜¿é‡Œå¤©æ± é¦–å±Šä¸æ–‡NL2SQL挑战赛top6
ayoungprogrammer/nlquery
Natural Language Engine on WikiData