On June 24, 2026, Google announced that it has natively integrated a "Computer Use" capability—letting the model autonomously operate on-screen interfaces—into its mainline Gemini 3.5 Flash model, scoring 78.4% on the real-environment OSWorld-Verified benchmark. Previously offered through standalone preview models such as those in the Gemini 2.5 line, the feature is now built into Flash, which keeps cost and speed low. Google describes it as a built-in tool "delivering our best performance yet for agentic computer use tasks."
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.