datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
sql-create-context
Overview
This dataset builds from WikiSQL and Spider.
There are 78,577 examples of natural language queries, SQL CREATE TABLE statements, and SQL Query answering the question using the CREATE statement as context. This dataset was built with text-to-sql LLMs in mind, intending to prevent hallucination of column and table names often seen when trained on text-to-sql datasets. The CREATE TABLE statement can often be copy and pasted from different DBMS and provides table names, column… See the full description on the dataset page: https://huggingface.co/datasets/b-mc2/sql-create-context.bird-critic-1.0-sqlite
📢 Update 2026-03-23
We release BIRD-Critic-SQLite, a dataset containing 500 high-quality user issues focused on real-world SQLite database applications. Along with the dataset, we also release three RL-trained models: BIRD-Talon-14B, BIRD-Talon-7B, and BIRD-Zeno-7B. The schema file is included in the code repository https://github.com/bird-bench/BIRD-CRITIC-1/blob/main/baseline/data/sqlite_schema.jsonl
BIRD-CRITIC-1.0-SQLite
BIRD-Critic is the first SQL debugging… See the full description on the dataset page: https://huggingface.co/datasets/birdsql/bird-critic-1.0-sqlite.bird_sql_dev_20251106
BIRD-SQL Dev
🆕 Update 2025-11-06
We would like to express our sincere gratitude to the community for their continuous support and constructive feedback on the BIRD-SQL Dev dataset. Over the past year, we have received valuable suggestions through GitHub discussions, emails, and user reports. Based on these insights, we organized a quality review program led by a team of five PhD researchers in Data Science and AI, supported by a globally distributed group of industry… See the full description on the dataset page: https://huggingface.co/datasets/birdsql/bird_sql_dev_20251106.livesqlbench-base-lite-sqlite
🚀 LiveSQLBench-Base-Lite
A dynamic, contamination‑free benchmark for evaluating LLMs on complex, real‑world text‑to‑SQL tasks.
🌐 LiveSQLBench Website • 🌐 BIRD-INTERACT Project Page • 📄 Paper • 💻 LiveSQLBench GitHub • 💻 BIRD-INTERACT GitHub
Maintained by the 🦜 BIRD Team @ HKU & ☁️ Google Cloud
📊 LiveSQLBench Overview
LiveSQLBench (BIRD-SQL Pro v0.5) is a contamination-free, continuously evolving benchmark designed to evaluate LLMs on complex, real-world… See the full description on the dataset page: https://huggingface.co/datasets/birdsql/livesqlbench-base-lite-sqlite.Text-to-sql-v1six-gym-sqlite
📢 Update 2026-03-23
We release BIRD-Critic-SQLite, a dataset containing 500 high-quality user issues focused on real-world SQLite database applications. This dataset is the train split of BIRD-Critic-SQLite, comprising 5,000 data instances for model training and development. Along with the dataset, we also release three RL-trained models: BIRD-Talon-14B, BIRD-Talon-7B, and BIRD-Zeno-7B.
📋 Dataset Structure
Below is a description of the dataset fields and additional… See the full description on the dataset page: https://huggingface.co/datasets/birdsql/six-gym-sqlite.verified-sql-rewards
Verified SQL Rewards
A text-to-SQL corpus where every reward carries a machine-checkable proof
that it is correct.
Questions, all independently verified
109,306
Databases
1,400 across 7 schema families
Tables / data rows
4,400 / ~19.6 million
Unique (question, answer) pairs
102,764
Candidates refused and published
12,150
Verification pass rate
90.00%
Trivial baseline (always answer 0)
1.83%
Each item is a natural-language question, a gold SQL query… See the full description on the dataset page: https://huggingface.co/datasets/rasinmuhammed/verified-sql-rewards.BIRD-SQL-TRAINbird-sql-mini-devSQL-API-Bench
Dataset Card for Dataset Name
This dataset contains QA that requires DB and API access at the same time. It is composed of two new benchmarks consisting of questions whose answers require a
combination of database and API calls, both of
which are augmentations of the popular Spider
dataset and benchmark.
Benchmark I replaces a fraction of the real Spider database tables with
equivalents that are executed via APIs. This allows us to directly test the mechanism by which
database and… See the full description on the dataset page: https://huggingface.co/datasets/ibm-research/SQL-API-Bench.Persian-Business-Text-to-SQL-Gold-1K
Persian Business Text-to-SQL Gold-1K
1,000 Persian-native, execution-verified business Text-to-SQL examples for fine-tuning and benchmarking.
مجموعهای ۱۰۰۰ نمونهای برای تبدیل درخواستهای فارسی کسبوکار به SQL، همراه با دیتابیسهای SQLite اجرایی، schema کامل، متادیتای سختی/مهارت و ارزیابی مبتنی بر Execution Accuracy.
Motivation
BIRD emphasizes database-grounded Text-to-SQL and execution accuracy; Spider 2.0 pushes toward realistic enterprise database workflows.… See the full description on the dataset page: https://huggingface.co/datasets/jumplander/Persian-Business-Text-to-SQL-Gold-1K.mirror-sql
MIRROR-SQL
Provenance-Controlled Database Environments for Text-to-SQL Agents.
13 PostgreSQL environments · 176 tables · 2762 columns · 390 annotated question/SQL pairs.
MIRROR-SQL takes the opposite approach to contamination from every other text-to-SQL corpus.
Spider and BIRD sample public databases. BEAVER uses real private warehouses that cannot be
redistributed. LiveSQLBench out-runs leakage temporally by rebuilding from changing sources.
MIRROR-SQL instead purpose-builds… See the full description on the dataset page: https://huggingface.co/datasets/1digitaldesign/mirror-sql.text-to-sql-shop
Text-to-SQL on a seeded store schema, with checkpoints
A question about an online store's database in, one PostgreSQL query out, graded by
a program: run the query, compare the result set to the gold query's result. The
schema (8 tables, seeded, schema.sql + seed.sql), the verifier, the trainer and
the benchmark runner are the
recipes/04-train/text-to-sql
recipe in the open-source whileai SDK.
Splits
config
rows
what
train
1764
tasks: question, gold_sql… See the full description on the dataset page: https://huggingface.co/datasets/while-ai/text-to-sql-shop.bird_sql_california_cot_v2av_sql_preprocessed_data
Dataset Card for Preprocessed Text-to-SQL Benchmarks
This repository contains preprocessed data for several text-to-SQL benchmarks, as presented in the paper AV-SQL: Decomposing Complex Text-to-SQL Queries with Agentic Views.
The official code for the AV-SQL framework can be found on GitHub: pminhtam/AV-SQL.
Dataset Summary
This repository contains preprocessed data for several text-to-SQL benchmarks:
BIRD
KaggleDBQA
Spider
sciencebenchmark
BEAVER
Spider2-Lite… See the full description on the dataset page: https://huggingface.co/datasets/griffith-bigdata/av_sql_preprocessed_data.enterprise-text-to-sql-benchmark
Enterprise Text-to-SQL Benchmark
3,045 natural-language questions paired with executable PostgreSQL, over a
12-table enterprise schema (sales, catalogue, logistics, HR).
Built to answer one question honestly: does fine-tuning actually improve
text-to-SQL? On this benchmark, a QLoRA fine-tune of Qwen3-8B took strict
execution accuracy from 10.82 % to 50.99 % — and the benchmark is designed
so that number cannot be inflated by leakage or by string-matching.
split
rows… See the full description on the dataset page: https://huggingface.co/datasets/hari-krishna-ai/enterprise-text-to-sql-benchmark.know_sqlplease use the val ign file for training, its much cleaner. thanks :)
GRAST-SQL-Spider
⚠️ This dataset has been deprecated. Please use the updated version below, which includes improved quality checks:https://huggingface.co/datasets/griffith-bigdata/GRAST-SL-evaluation-set
GRAST-SQL Spider Training & Evaluation Dataset
This dataset is processed from the original Spider dataset, with extracted schema information and used_columns from SQL queries. It is prepared solely for training, and evaluating schema filtering in the context of the GRAST-SQL paper.… See the full description on the dataset page: https://huggingface.co/datasets/griffith-bigdata/GRAST-SQL-Spider.SQLFlow
Text2SQL-Flow Dataset Repository
This repository contains the SQLFlow dataset.
The SQLFlow dataset is a large-scale, high-quality collection of semantically valid and structurally diverse Text-to-SQL examples, generated using a comprehensive SQL-aware data augmentation framework.
For more details, please visit the GitHub repository:🔗 https://github.com/TechNomad-ds/Text2SQL-Flow
sql-dataSQLDatabird_sql_superhero_cot_methodsql-create-context-copy
Fork of b-mc2/sql-create-context
Overview
This dataset builds from WikiSQL and Spider.
There are 78,577 examples of natural language queries, SQL CREATE TABLE statements, and SQL Query answering the question using the CREATE statement as context. This dataset was built with text-to-sql LLMs in mind, intending to prevent hallucination of column and table names often seen when trained on text-to-sql datasets. The CREATE TABLE statement can often be copy and pasted from… See the full description on the dataset page: https://huggingface.co/datasets/philschmid/sql-create-context-copy.CSpider_and_DUSQL_sql_create_context此数据集原始数据集是合并CSpider和DUSQL。根据b-mc2/sql-create-context所创建。
Effi-SQL
Effi-SQL
Update 2026-06-12
We release Effi-SQL, a dataset suite for SQL efficiency optimization.
This collection includes:
Effi-SQL Benchmark: a benchmark for evaluating SQL efficiency optimization methods.
Diff-SQL Training Dataset: training data used by Diff-SQL, including data for the Patch Generator and Constraint Aligner.
Dataset Fields
Effi-SQL Benchmark
id: A unique identifier for each benchmark instance.
db: The database… See the full description on the dataset page: https://huggingface.co/datasets/birdsql/Effi-SQL.sql-qa-chinooksql-create-context-id
Overview
This dataset is a fork from sql-create-context
This dataset builds from WikiSQL and Spider.
There are 78,577 examples of natural language queries, SQL CREATE TABLE statements, and SQL Query answering the question using the CREATE statement as context. This dataset was built with text-to-sql LLMs in mind, intending to prevent hallucination of column and table names often seen when trained on text-to-sql datasets. The CREATE TABLE statement can often be copy and pasted from… See the full description on the dataset page: https://huggingface.co/datasets/detakarang/sql-create-context-id.CSpider_sql_create_context此数据集原始数据集是CSpider。根据b-mc2/sql-create-context所创建。
sql-create-context-spider-intersecttext-to-sql-eval-predictions
What the text-to-SQL models actually generated
Every prediction behind the numbers in
qwen3-8b-text2sql-qlora: the 453 test
questions of the enterprise text-to-SQL benchmark,
each answered by four configurations of the same model, each answer executed against the reference
PostgreSQL database and scored by comparing result sets. 1,812 rows.
I published this because the headline table (10.82 % → 50.99 % → 52.10 %) is the least interesting part of
that project. The interesting… See the full description on the dataset page: https://huggingface.co/datasets/hari-krishna-ai/text-to-sql-eval-predictions.