Google has launched Gemini 3.7 Flash, shortly after releasing Gemini 3.6 Flash, with the company saying the new model offers faster coding and AI agent capabilities. Google is also offering an introductory token promotion with more than half off the regular price.Gemini 3.7 Flash builds on Gemini 3.6 Flash by incorporating feedback from developers to improve its performance, particularly in coding tasks such as debugging, code accuracy and the quality of production-ready code, while requiring fewer instructions.
On the WebDev Arena benchmark, Gemini 3.7 Flash scored an Elo rating of 1,588, compared with 1,538 for Gemini 3.6 Flash.
For tasks that require large datasets, including finance, law and life sciences, Gemini 3.7 Flash is also said to provide stronger analysis and reasoning while delivering more accurate information.
Google is offering an introductory promotion for Gemini 3.7 Flash at nearly half the price, which will run until the end of 2026. Input tokens are priced at $0.75 per 1 million tokens, while output tokens cost $3.75 per 1 million tokens.
Google also announced that Gemini Spark will switch to Gemini 3.7 Flash for users of the Google AI Ultra and AI Pro plans in supported countries, allowing AI agents to operate more continuously, quickly and smoothly.
