Google Unveils Gemini 3.7 Flash with Improved Coding Capabilities

Google has rolled out Gemini 3.7 Flash, an upgraded version of its AI model that puts coding and agent workflows at its core. The new model boasts improved adaptability when encountering roadblocks, better clarifies intent when necessary, and follows instructions with greater fidelity.

The launch marks a rapid iteration on Google's Flash line, coming just three weeks after the release of Gemini 3.6 Flash. This brief turnaround is attributed to developer feedback and algorithmic improvements. For enterprise developers, the more significant story may be the combination of intelligence gains with lower inference costs: through the end of 2026, Gemini 3.7 Flash costs $0.75 per million input tokens and $3.75 per million output tokens.

Google's benchmarks show a substantial generational improvement in several software engineering tests. On FrontierCode 1.1 Main, which measures production code quality, Gemini 3.7 Flash scores 43.6%, up from 34.4% for Gemini 3.6 Flash. This also narrowly exceeds the 42.7% reported for Claude Sonnet 5 and 41.3% for GPT-5.6 Terra.

The introduction of Gemini 3.7 Flash underscores Google's efforts to embed its model into enterprise workflows. The company has made its model available through the Gemini Enterprise Agent Platform and Gemini Enterprise app, with a temporary price cut in place until December 31st.