Skip to content

[ENHANCEMENT] Enable browser-use for all image-capable models #8116

@hannesrudolph

Description

@hannesrudolph

Type

Enhancement

Problem / Value

Models that can accept images don’t automatically get computer-use capability, so the browser-use tool may be unavailable even in modes that allow it. Making image support imply computer use creates a predictable, simpler experience and unlocks browser-based verification and research flows across more models.

Context

Affects anyone choosing an image-capable model and expecting consistent tool access. Currently, computer use is gated by a separate capability flag that isn’t always present for vision models across providers. As a result, the browser-use tool can be hidden even though the model handles images. This change should auto-enable browser-use whenever:

  • The selected model supports images, and
  • The current mode allows browser tools, and
  • Browser tools are enabled in settings.

Constraints/Preferences

  • Apply globally: image support implies computer use.
  • Keep existing mode and settings gating intact (mode must allow browser and browser tools must be enabled).

Metadata

Metadata

Assignees

No one assigned

    Labels

    EnhancementNew feature or requestIssue - Needs ApprovalReady to move forward, but waiting on maintainer or team sign-off.

    Type

    No type

    Projects

    Status

    Done

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions