GroundUI-1K

multimodal official site →

A subset of GroundUI-18K for UI grounding evaluation, where models must predict action coordinates on screenshots based on single-step instructions across web, desktop, and mobile platforms.

Methodology

Imported from llm-stats public benchmark metadata. Modality: multimodal. Max score: 1. Categories: grounding, multimodal, vision. Language: en. Verified by llm-stats: no.

Leaderboard

  1. Nova Pro self-reported llm-stats
    81.4%
  2. Nova Lite self-reported llm-stats
    80.2%