Google’s Gemini 3.7 Flash targets coding and agents with a 50% introductory price cut

Google has launched Gemini 3.7 Flash, a new version of its AI model that prioritizes coding, agentic workflows, and knowledge work. The company cut API prices in half, temporarily reducing the cost to $0.75 per million input tokens and $3.75 per million output tokens, through the end of 2026. The price increases to $1.50 per million input tokens and $7.50 per million output tokens starting Jan.

1, 2027. The launch comes just three weeks after Gemini 3.6 Flash, with the speed driven by developer feedback and algorithmic improvements.

Google highlights Gemini 3.7 Flash's improvements in coding and agent workflows, stating that the model is better at adapting when encountering roadblocks, clarifying intent, and executing multi-step plans more reliably. These enhancements allow for fewer human interventions in enterprise coding and business agents, potentially reducing total operating costs for teams deploying high-volume coding and business agents.

The model shows notable gains in coding and agent workloads. In software engineering tests, Gemini 3.7 Flash scores 43.6% on FrontierCode 1.1 Main, up from 34.4% for Gemini 3.6 Flash, narrowly surpassing Claude Sonnet 5's 42.7% and GPT-5.6 Terra's 41.3%. In web development, Gemini 3.7 Flash receives an Elo score of 1588 on Code Arena, outperforming Claude Sonnet 5's 1538 and GPT-5.6 Terra's 1541.

Beyond software development, Gemini 3.7 Flash shows significant improvements in enterprise workflow automation. On AutomationBench, the model scores 30.4%, a sharp increase from 17.0% for Gemini 3.6 Flash, outperforming Claude Sonnet 5's 10.7% and GPT-5.6 Terra's 23.6%. This combination of gains makes Gemini 3.7 Flash more competitive in enterprise agent workloads, where reliability across multiple tasks is crucial.