live wire
nl2sql.ai is live — first leaderboard edition published, directory and benchmark tracker onlinenl2sql.aiBIRD leaderboard: top cluster holds in the low-to-mid 70s on dev execution accuracy; human reference 92.96bird-bench.github.ioSpider 2.0 remains the wall: launch-paper agentic baseline solved ~17% of enterprise tasksarXivDirectory day one: 12 systems catalogued across cloud-native, OSS, and research categoriesnl2sql.aiVanna remains the most-starred OSS text-to-SQL framework; RAG-on-your-own-pairs still the default patternGitHubWren AI ships steadily on its MDL semantic layer — the OSS counterpart to vendor semantic modelsGitHubUber QueryGPT post remains the canonical enterprise-scale case study: routing beats generation at 1000s of tablesUber EngineeringWatch item: semantic-layer interop — every platform has one, none of them talk to each otheranalysis deskDB-GPT community keeps shipping: fine-tuning hub and AWEL workflows anchor the self-hosted stackGitHubDesk assignments filed: releases and papers, leaderboards, the directory. Cadence: continuousnl2sql.ainl2sql.ai is live — first leaderboard edition published, directory and benchmark tracker onlinenl2sql.aiBIRD leaderboard: top cluster holds in the low-to-mid 70s on dev execution accuracy; human reference 92.96bird-bench.github.ioSpider 2.0 remains the wall: launch-paper agentic baseline solved ~17% of enterprise tasksarXivDirectory day one: 12 systems catalogued across cloud-native, OSS, and research categoriesnl2sql.aiVanna remains the most-starred OSS text-to-SQL framework; RAG-on-your-own-pairs still the default patternGitHubWren AI ships steadily on its MDL semantic layer — the OSS counterpart to vendor semantic modelsGitHubUber QueryGPT post remains the canonical enterprise-scale case study: routing beats generation at 1000s of tablesUber EngineeringWatch item: semantic-layer interop — every platform has one, none of them talk to each otheranalysis deskDB-GPT community keeps shipping: fine-tuning hub and AWEL workflows anchor the self-hosted stackGitHubDesk assignments filed: releases and papers, leaderboards, the directory. Cadence: continuousnl2sql.ai
nl2sql.ai

open-source · since 2023

Vanna

Vanna.AI

Popular open-source Python framework for text-to-SQL using retrieval-augmented generation: train on DDL, documentation, and example query pairs; works across many LLMs, vector stores, and databases.

ragpythonframework
desk notes

One of the most-starred text-to-SQL projects on GitHub. The train-on-your-own-pairs loop made RAG-over-schema the default OSS pattern.

leaderboard history
  • 2026-08#479.5The default OSS starting point: RAG over your own DDL and query pairs, any LLM, any store.
Vanna — nl2sql.ai