Home
News
Details

Google's Gemini-SQL2 achieves 80.04% accuracy on the BIRD benchmark.

iconKuCoinFlash
Share
Share IconShare IconShare IconShare IconShare IconShare IconCopy
AI summary iconSummary

expand icon
On-chain news: On June 12, Google Research unveiled Gemini-SQL2, achieving 80.04% accuracy on the BIRD benchmark. Built on Gemini 3.1 Pro, it improves by 2.9 points over the previous version. BIRD evaluates over 12,700 real-world SQL scenarios. The model will power tools such as BigQuery Studio and Cloud SQL Studio. Real-world assets (RWA) news: The update aims to simplify database access for non-technical users through natural language.
ME AI News: According to monitoring by Beating, Google Research released the text-to-SQL technology Gemini-SQL2 on June 12. The new system, built on the Gemini 3.1 Pro model, achieved an execution accuracy of 80.04% on the BIRD benchmark, a leading industry evaluation for single-model text-to-SQL systems, setting a new performance record. The earlier version, Gemini-SQL, powered by Gemini 2.5 Pro and utilizing multi-task supervised fine-tuning, previously led the leaderboard with an execution accuracy of 77.14%. The new version improved this by 2.9 percentage points. BIRD, as the industry standard for evaluating text-to-SQL systems, includes over 12,700 question-SQL pairs, testing real-world database scenarios such as multi-table joins, dirty data handling, and complex business context reasoning, with accuracy measured by whether the generated SQL executes correctly in the database. Gemini-SQL2 is not being released as a standalone large model but is being promoted as an underlying database query capability. Google plans to gradually integrate this functionality into its existing data services, including BigQuery Studio, AlloyDB AI, and Cloud SQL Studio, enabling non-technical users to query enterprise databases directly through natural language. (Source: BlockBeats)
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information. Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.