← Models

Computer Use Preview

Google · Gemini Computer Use

Google's preview computer-use model surface for browser and interface control workflows.

Part of Gemini Pro family · Other versions: Gemini 3.1 Pro Preview , Gemini 2.5 Pro TTS Preview , Gemini 3 Pro Preview , Gemini 2.5 Pro
Type
multimodal
Context
128K tokens
Max Output
N/A
Status
preview
Input
$0.9/1M tok
Output
$3.5/1M tok
API Access
Yes
License
proprietary
computer-use browser-control agentic multimodal preview
Released October 2025 · Updated April 4, 2026

Overview

Freshness note: Model capabilities, limits, and pricing can change quickly. This profile is a point-in-time snapshot last verified on April 4, 2026.

Computer Use Preview is Google’s preview browser-control model surface for Gemini API workflows. It matters because it makes browser and UI interaction an explicit model-level capability rather than only a higher-level product feature layered on top of general language models.

Capabilities

This model class is aimed at workflows where an agent must inspect a page, reason about interface state, and take bounded actions such as clicking, filling fields, navigating, or collecting evidence. That makes it relevant for task execution across legacy tools, internal dashboards, and operational web workflows where APIs are incomplete or unavailable.

Technical Details

Google’s current Gemini API pricing docs list Computer Use Preview with a 128,000-token text context window. The current public pricing surface does not make the max output limit as clear as it does for the Flash line, so Signal Lens stores maxOutput as 0 and treats it as N/A until Google exposes a cleaner public spec.

Google’s current model catalog also ties computer-use capabilities to the broader Gemini model surface, which suggests Google is treating browser control as a growing first-class agent feature rather than a one-off experiment.

Pricing & Access

Google’s current Gemini API pricing lists Computer Use Preview at:

  • Input: $0.90 per 1M text tokens
  • Output: $3.50 per 1M tokens

Availability is through the Gemini API preview surface. Teams should assume faster behavior changes and stricter evaluation needs than on stable production models.

Best Use Cases

Use Computer Use Preview for supervised browser automation, internal operations tooling, form-based workflows, and evaluation environments where a reviewer can approve actions before systems of record are changed. It is not a good fit for unsupervised high-risk automation.

Comparisons

  • Gemini 3 Flash (Google): Broader fast multimodal model, while Computer Use Preview is the more specific browser-control route.
  • Perplexity Computer (Perplexity): Packaged browser-agent product for end users, while Computer Use Preview is a model-layer building block.
  • OpenClaw-style stacks: Self-hosted alternative when teams want tighter control over the runtime and tool chain.