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

Welcome to nl2sql.ai

A trade publication for one technical beat: systems that turn natural language into SQL. Continuous coverage, evidence-diffed rankings, public corrections.

By The Editorial Desk· Aug 10, 2026

nl2sql.ai covers one beat, relentlessly: natural language to SQL. New model releases, benchmark movements, open-source projects, enterprise platform features, research papers, and the occasional post-mortem when things go wrong.

What we publish

  • The wire — a continuously updated feed of short, sourced items: releases, papers, leaderboard moves, announcements.
  • Stories — reported pieces when events deserve more than a headline.
  • The leaderboard — a monthly, evidence-diffed ranking of the systems that matter.
  • The directory — reference entries for every serious system on the beat, kept current.
  • The benchmark tracker — BIRD, Spider 2.0, and whatever credible evaluation comes next, with every number linked to its source.

The rules we publish under

  1. Every claim cites a source. Wire items and stories link to the primary artifact — the release notes, the paper, the leaderboard entry.
  2. Rankings are evidence-diffed. Score changes name the evidence that moved them. No pay-for-placement, ever.
  3. Corrections are public. Corrected or retracted stories stay visible with a note explaining what changed.

Why this beat

Text-to-SQL sits at the exact pressure point of enterprise AI: it is the interface between "everyone can ask questions of data" and "the SQL was subtly wrong and nobody noticed." Benchmarks like Spider 2.0 show frontier systems still failing most realistic enterprise tasks, while every major data platform ships an NL2SQL feature anyway. That gap — between demo and dependable — is a news beat. We intend to cover it relentlessly.

Tips, corrections, and challenges: desk@nl2sql.ai. The wire is live.

Filed by The Editorial Desk. Corrections: desk@nl2sql.ai · Our standards →
more from the desk