Google Research's Gemini-SQL2 achieves 80% accuracy on text-to-SQL benchmark, outpacing OpenAI and Anthropic
By
Matthias Bastian
A second-rack bagel that's nearly first-rack. Tasty stuff.
Summary
Google Research has unveiled Gemini-SQL2, a text-to-SQL system built on Gemini 3.1 Pro that translates natural language into executable SQL database queries. It leads the BIRD benchmark with an execution accuracy of 80.04%, significantly outperforming OpenAI's GPT-5.5-xhigh (72.8%) and Anthropic's Claude Opus 4.6 (70.9%). Google suggests the technology could enhance natural language features across its data services.
Key quotes
· 3 pulledGemini-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.
Google says the technology could improve natural language features across its data services.
You might also wanna read
Google Releases Gemini 3.1 Pro AI Model with 2M Token Context for Complex Tasks
Google has released Gemini 3.1 Pro, an advanced AI model designed for complex, multi-step tasks that require sophisticated reasoning and ana
blog.google·3mo ago
Gemini 3.1 Pro Benchmark Performance Analysis Across Multiple AI Evaluation Tasks
The article presents benchmark performance data for Gemini 3.1 Pro, comparing it against other leading AI models including Gemini 3 Pro, Son
deepmind.google·3mo ago
Google Releases Gemini 3 Pro AI Model with Audio Transcription and New Benchmark Performance
Google has released Gemini 3 Pro, an upgraded version of Gemini 2.5 that brings it to parity with leading rival AI models. The article provi
Author's Preference for Google Gemini AI Models Over Competitors
The article discusses the author's preference for Google's Gemini AI models over competitors like Anthropic and OpenAI, praising Gemini's sp
robinsloan.com·3mo ago
Google's Gemini 3 AI Model Tops Benchmarks and Leaderboards, Outperforming Competitors
Google's Gemini 3 AI model has been released to widespread acclaim, topping benchmarks and leaderboards while outperforming competitors like
The Verge·6mo ago
Google Unveils Gemini: A Multimodal AI Model to Rival GPT-4
Google's Gemini is introduced as its largest and most capable AI model, designed to be multimodal and capable of understanding and combining
Product Hunt·10mo ago