Gemini 3.5 Flash Adds Native Computer Use Tool
Gemini 3.5 Flash now includes a built-in computer use tool, enhancing its ability to build agents that can see, reason, and act across different environments. This integration aims to improve performance for long-horizon and enterprise automation tasks, making it easier for developers and enterprises to build custom agents. The feature is available via the Gemini API and Gemini Enterprise Agent Platform, with enhanced safety measures for sensitive actions.
- →Gemini 3.5 Flash integrates computer use tool
- →Enterprise safeguards for agent safety
- →Improved performance for automation tasks
- →Availability and getting started
Security (1) ›
- Enterprise safeguards for agent safety
New optional enterprise safeguard systems are available, including explicit user confirmation for sensitive actions and automatic task stoppage upon detection of indirect prompt injection. These are part of a defense-in-depth approach alongside sandboxing and access controls.
Features (1) ›
- Gemini 3.5 Flash integrates computer use tool
Gemini 3.5 Flash now natively supports computer use, previously a standalone tool, for enhanced agentic tasks. This allows developers to build custom agents capable of interacting with browser, mobile, and desktop environments.
Enhancements (1) ›
- Improved performance for automation tasks
The integration of computer use in 3.5 Flash is designed to improve performance for long-horizon and enterprise automation tasks, such as continuous software testing and knowledge work within professional applications.
Notes (1) ›
- Availability and getting started
Computer use in Gemini 3.5 Flash is accessible through the Gemini API and Gemini Enterprise Agent Platform. A demo environment hosted by Browserbase and a reference implementation are available for developers to start building.
https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-computer-use-gemini-3-5-flash/