Codesota · Natural Language Processing · Language Modeling · Bird-SQL (dev)Tasks/Natural Language Processing/Language Modeling
Language Modeling · benchmark dataset · EN

BIRD-SQL (BIg Bench for Large-Scale Database-Grounded Text-to-SQLs).

Development split (dev) of BIRD-SQL (BIRD). BIRD-SQL is a large cross-domain text-to-SQL benchmark designed to evaluate natural-language-to-SQL parsing against realistic, value-rich relational databases. BIRD contains 12,751 text-to-SQL question–SQL pairs grounded on 95 databases (total ~33.4 GB) spanning ~37 professional domains; it emphasizes database values (dirty/noisy values and external-knowledge grounding) to better match real-world DB assistant scenarios. The benchmark provides standard splits including a development (dev) split (the dev archive is distributed by the authors) which is commonly used for model evaluation (accuracy / execution metrics in papers). The evaluation result referenced corresponds specifically to the development split as reported in Table 6 (metric: accuracy).

Paper Submit a result
§ 01 · Leaderboard

Best published scores.

No results indexed yet — be the first to submit a score.

No benchmark results indexed yet
§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies