The release comes about three weeks after Gemini 3.6 Flash and follows feedback from developers, Google said in a blog post dated Aug. 13. The company described the new model as its most intelligent “workhorse” model so far for coding and agent-based workflows.
AI agents are systems that can perform a sequence of tasks rather than simply answer a single question. For example, an agent may be able to read files, use software tools, draft documents and carry out several steps toward a goal with less human intervention.
Google said Gemini 3.7 Flash improves on its predecessor in software engineering, web development and knowledge-intensive work such as finance, law and biosciences. The company said the model is better at debugging and resolving coding issues, following instructions and planning multi-step actions.
In Google's published evaluations, Gemini 3.7 Flash scored 43.6% compared with 34.4% for Gemini 3.6 Flash on FrontierCode 1.1 Main, while its score on DeepSWE v1.1 was 65.3% versus 49.0% for the previous model.
The model also recorded a higher score on Arena.ai's WebDev Arena, with an Elo rating of 1,588 compared with 1,538 for Gemini 3.6 Flash, according to Google. On the GDP.pdf benchmark for complex document processing, it scored 34.0% compared with 22.0%, while AutomationBench results were 30.4% versus 17.0%.
Google is also positioning the model as a lower-cost option for developers. Through the end of 2026, Gemini 3.7 Flash will cost $0.75 per 1 million input tokens and $3.75 per 1 million output tokens. Google said the introductory pricing ends on Dec. 31, after which prices will rise to $1.50 and $7.50, respectively.
The new model is being made available across Google's developer and enterprise platforms, including the Gemini API, Google AI Studio, Android Studio, Google Antigravity, Gemini Enterprise Agent Platform and the Gemini Enterprise app.
Google is also bringing Gemini 3.7 Flash to Gemini Spark, its personal AI agent for Google AI Pro and Ultra subscribers in supported markets. Google said Spark will use the updated model for tasks involving Google Workspace applications, including consolidating files, drafting emails and updating status documents.
The company said the release also includes updated safeguards addressing potential misuse in areas including chemical, biological, radiological and nuclear risks, as well as cyber-related misuse.
The launch comes as major AI companies compete to improve models that can move beyond generating text to performing longer, more complicated tasks. For users and businesses, Google's emphasis on coding, tool use, reliability and lower operating costs reflects the growing push to make AI systems useful in everyday software and workplace workflows.
I truly appreciate you spending your valuable time here. To help make this blog the best it can be, I would love your feedback on this post. Let me know in the comments: How could this article be better? Was it clear? Did it have the right amount of detail? Did you notice any errors?
If you found any of the articles helpful, please consider sharing it.