Google Research's Gemini-SQL2 tops text-to-SQL benchmarks by a wide margin

Ad

Skip to content

The Decoder

AI, Menschen, Wirtschaft

-->

Sign In

Register

Subscribe Now

The Decoder

Opens discord in a new tab<br>Opens LinkedIn in a new tab

Google Research's Gemini-SQL2 tops text-to-SQL benchmarks by a wide margin

Matthias Bastian

View the LinkedIn Profile of Matthias Bastian

Jun 13, 2026

Google Research unveiled Gemini-SQL2, a new text-to-SQL system built on Gemini 3.1 Pro. It translates natural language into executable SQL database queries. On the BIRD benchmark, which measures how accurately these translations work, Gemini-SQL2 hits an execution accuracy of 80.04 percent, putting it in first place, according to Google. OpenAI's GPT-5.5-xhigh scores about 72.8 percent, and Anthropic's Claude Opus 4.6 lands around 70.9 percent. Models from Databricks, AWS, Tencent, and Alibaba all trail well behind.

Gemini-SQL2 leads the BIRD text-to-SQL leaderboard with 80.04 percent execution accuracy, outpacing competitors from OpenAI, Anthropic, Databricks, and others. | Image: Google<br>Google Research points out that turning natural language into correct SQL is especially hard because data is often layered and queries need to account for complex business logic. The generated SQL queries both look correct and execute successfully, the company says.

Better SQL understanding could improve natural language features across Google's data services more broadly, according to Google. The research team hasn't said anything about a public release of the model, and there's no paper yet either.Ad<br>DEC_D_Incontent-1

Ad

AI News Without the Hype – Curated by Humans

Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.

Subscribe now

Source: Google Research

Ask about this article…

Search

Top<br>Stories

Landmark German ruling declares Google's AI Overviews are Google's own words and makes it liable for false answers

Anthropic releases Claude Fable 5 and Mythos 5 with major gains in coding and science

Researchers pinpoint why larger language models pick up skills that small ones miss

OpenAI now says "entirely automating everything is not the future we want"

OpenAI says "chat is dead" and plans to rebuild ChatGPT as a full-blown agent app

Don't Miss<br>What Matters

Stay in the loop on AI. Clear, useful, no fluff.

Your email address

Subscribe to our Newsletter

Most<br>Popular

Landmark German ruling declares Google's AI Overviews are Google's own words and makes it liable for false answers

For $1.3 million a month, OpenClaw founder Peter Steinberger runs 100 AI agents that code, review PRs, and find bugs

Google launches a tiny board that runs Gemma 3 locally

Fields Medalist says ChatGPT 5.5 Pro delivered "PhD-level" math research in under two hours with zero human help

Researchers let Claude Code discover AI scaling algorithms that humans probably wouldn't have designed

Claude Mythos reportedly solves OpenAI's landmark Erdős problem with a "cute, simple proof"

AI Community<br>& Insights

The Decoder

Follow The Decoder for AI news, background stories and expert analyses.

Opens LinkedIn in a new tab

Opens Discord in a new tab

BETA-TEST

&times;

Start new search

&times;

Send

wpDiscuz

Insert

BETA-TEST

&times;

Start new search

&times;

Send

wpDiscuz

Insert