kapynAI / Models

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

Gemini-SQL2 sets new text-to-SQL benchmark standard. This Google Research model, built on Gemini 3.1 Pro, achieved 80.04% accuracy on the BIRD benchmark, significantly outperforming competitors like OpenAI and Anthropic. This advancement promises to enhance natural language capabilities across Google's data services.

The Decoder·Jun 13, 2026

Opening Kapyn…

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