Gemini 2.5 Computer Use Preview 10-2025
gemini-2.5-computer-use-preview-10-2025
Google · chat · api · seen 14m ago
Preview
Alert me on changes
Context
131.1K
Max output
65.5K
Input $/1M
$1.25
Output $/1M
$10.00
Modalities
text
Released
01 Oct 2025
AI summary
● machine-written
Google releases Gemini 2.5 Computer Use Preview for UI automation
Gemini 2.5 Computer Use Preview is an agentic model designed for direct interface interaction, enabling visual reasoning and precise cursor control. It supports browser automation, multi-step software tasks, and UI testing. The model is available via the Gemini API in preview status with a 128K token input limit and 64K output limit.
What's new
- Supports computer use capabilities for digital environment navigation
- Designed for browser automation and UI testing with visual reasoning
- Context window of 131K tokens
- Available in preview status as of October 2025
Best for
Browser automation and web navigationMulti-step software tasks requiring visual reasoningUI testing with precise cursor controlAgentic interface interaction workflows
Source: https://ai.google.dev/gemini-api/docs/models/gemini-2.5-computer-use-preview-10-2025