Google Launches Gemini 3.7 Flash for Coding and AI Agent Workflows

Key Highlights of NEWS

• Google has launched Gemini 3.7 Flash, its latest AI model for coding, software engineering, web development, and AI agents.
• The model improves debugging, code generation, web development, document understanding, and business workflow automation.
• Google is offering Gemini 3.7 Flash at $0.75 per million input tokens and $3.75 per million output tokens through the end of 2026.
• Gemini 3.7 Flash is also rolling out to GitHub Copilot across several development environments.

Want to understand how AI models like Gemini 3.7 Flash are changing coding and AI workflows? Explore our Artificial Intelligence course to build a strong foundation in modern AI technologies and applications. 

Google has launched Gemini 3.7 Flash on August 13, 2026. The new AI model focuses on coding, software engineering, web development, and AI agent workflows. Google calls it its most intelligent workhorse model yet for coding and agents. The launch comes only three weeks after Gemini 3.6 Flash.

Google says the new model can handle complex tasks with better planning, tool use, and instruction following. It is also designed to reduce manual intervention during AI-powered workflows.

What Is Gemini 3.7 Flash?

Gemini 3.7 Flash is Google's latest Flash AI model for coding and AI agent workflows. It is designed to handle tasks that require multiple steps. These include debugging, software development, web development, and business automation.

The model can plan actions and use software tools. It can also adapt when a task encounters problems. Google says it can ask for clarification when needed. Gemini 3.7 Flash also targets enterprise workflows. Google says it performs better in areas such as finance, law, and biosciences.

The model improves on Gemini 3.6 Flash across several benchmarks. On FrontierCode 1.1 Main, it scored 43.6%, compared with 34.4% for Gemini 3.6 Flash. On DeepSWE v1.1, it scored 65.3%, compared with 49% for the previous model. For web development, Gemini 3.7 Flash also showed higher performance. Its WebDev Arena Elo score reached 1,588. Gemini 3.6 Flash scored 1,538 on the same benchmark.

How Does Gemini 3.7 Flash Improve AI Coding and Agents?

Gemini 3.7 Flash is designed to complete longer and more complex coding workflows. Google says the model is better at debugging, resolving software issues, and generating production-ready code. It also performs better when developers need an AI system to use tools across multiple steps.

The model also improves web development. It can create more functional layouts and feature-complete applications with fewer prompts. It can also reproduce designs using screenshots, images, or design systems as references.

Another focus is business workflow automation. On the AutomationBench benchmark, Gemini 3.7 Flash scored 30.4%. Gemini 3.6 Flash scored 17%. The model also improved on complex document understanding. It scored 34% on the GDP.pdf benchmark, compared with 22% for Gemini 3.6 Flash.

These improvements are important for AI agents. AI agents need to plan tasks, call tools, respond to problems, and complete workflows. Gemini 3.7 Flash is designed around these requirements.

Google is also using the model to power Gemini Spark. Spark is Google's personal AI agent for Google AI Pro and Ultra subscribers. With Gemini 3.7 Flash, Spark can consolidate files, draft emails, and update status documents using Google Workspace tools.

As AI moves from simple chatbots to coding agents and intelligent workflows, building reliable AI products requires deeper technical expertise. The IIT Kharagpur Executive Post Graduate Certificate in Building AI Products, Systems & Services can help professionals develop skills to design and build production-ready AI solutions. 

Where Is Gemini 3.7 Flash Available?

Gemini 3.7 Flash is available through several Google developer and enterprise products. Developers can access the model through the Gemini API, Google AI Studio, and Android Studio. Enterprises can use it through the Gemini Enterprise Agent Platform and Gemini Enterprise.

The model is also rolling out to GitHub Copilot. GitHub says Gemini 3.7 Flash improves web and app development, agentic coding, code quality, codebase research, and verification.

It is available to Copilot Pro, Pro+, Max, Business, and Enterprise users.

The model can be selected in Visual Studio Code, Visual Studio, Copilot CLI, GitHub Copilot cloud agent, and the GitHub Copilot app. It is also available across JetBrains, Xcode, and Eclipse. The rollout is gradual.

Google is also focusing on cost. Gemini 3.7 Flash costs $0.75 per million input tokens and $3.75 per million output tokens at its introductory rate through the end of 2026.

That is half the original Gemini 3.6 Flash pricing. The lower price could be important for companies running AI agents at scale. Agentic systems can make many model calls during a single workflow.

Google's strategy is therefore focused on both performance and cost efficiency.

Conclusion

Gemini 3.7 Flash strengthens Google's focus on coding and agentic AI. The model improves software engineering, web development, and business automation. Its lower introductory pricing also targets large-scale AI agent deployments.

Its arrival in GitHub Copilot expands access for developers. The model's use in Gemini Spark also shows Google's push toward AI systems that can complete tasks, not just answer questions.