Google debuts Gemini 3.7 Flash with deep price cuts

Google has introduced Gemini 3.7 Flash, a rapid update arriving only three weeks after the release of Gemini 3.6 Flash. This new model is designed as a highly optimized tool aimed at improving coding and agentic workflows. The release comes as Google's flagship Gemini 3.5 Pro remains delayed past its promised June launch window, suggesting the tech giant is leaning on iterative Flash updates to maintain momentum against competitors.

The new model shows notable performance gains across several key evaluations. On the FrontierCode 1.1 Main test, Gemini 3.7 Flash scored 43.6 percent, up from 34.4 percent on the previous version. Its DeepSWE v1.1 score rose from 49 percent to 65.3 percent, while its WebDev Arena score ticked up to 1,588 from 1,538. For document processing, the model scored 34 percent on the GDP.pdf benchmark compared to 3.6 Flash's 22 percent. Additionally, its AutomationBench score for business workflows jumped from 17 percent to 30.4 percent.

To attract developers, Google is offering Gemini 3.7 Flash at an introductory rate through the end of the year. Input tokens are priced at $0.75 per million and output tokens at $3.75 per million, representing a 50 percent discount compared to Gemini 3.6 Flash. However, this pricing still trails OpenAI's budget options; the rival's GPT 5.6 Luna model costs just $0.20 per million input tokens and $1.20 per million output tokens.

For practitioners, the model is immediately accessible via the Gemini API, AI Studio, and Gemini Enterprise. Individual users can access Gemini 3.7 Flash through the Gemini Spark agent in the Gemini app, though this is restricted to AI Pro or Ultra subscribers. The standard, free chatbot interface continues to run on Gemini 3.6 Flash for the time being.