Skip to main content

ScreenSpot Pro

A high-resolution GUI grounding benchmark for professional computer-use environments.

Benchmark score on ScreenSpot Pro — April 10, 2026

BenchLM mirrors the published score view for ScreenSpot Pro. GPT-5.4 leads the public snapshot at 85.4% , followed by Gemini 3.1 Pro (84.4%) and Muse Spark (84.1%). BenchLM does not use these results to rank models overall.

12 modelsMultimodal & GroundedCurrentDisplay onlyUpdated April 10, 2026

The published ScreenSpot Pro snapshot is tightly clustered at the top: GPT-5.4 sits at 85.4%, while the third row is only 1.3 points behind. The broader top-10 spread is 26.5 points, so the benchmark still separates strong models even when the leaders cluster.

12 models have been evaluated on ScreenSpot Pro. The benchmark falls in the Multimodal & Grounded category. This category carries a 12% weight in BenchLM.ai's overall scoring system. ScreenSpot Pro is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About ScreenSpot Pro

Year

2025

Tasks

GUI grounding tasks

Format

Interface element localization

Difficulty

Professional GUI grounding

ScreenSpot Pro matters for computer-use agents because it tests whether a model can find and ground the right UI target before it ever clicks or types. Strong GUI grounding is a prerequisite for reliable desktop agents.

BenchLM freshness & provenance

Version

ScreenSpot Pro 2025

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Benchmark score table (12 models)

1
85.4%
2
84.4%
3
84.1%
4
83.1%
5
72.7%
6
70.6%
7
68.2%
8
66.1%
9
65.6%
10
58.9%
11
57.2%
12
45.7%

FAQ

What does ScreenSpot Pro measure?

A high-resolution GUI grounding benchmark for professional computer-use environments.

Which model scores highest on ScreenSpot Pro?

GPT-5.4 by OpenAI currently leads with a score of 85.4% on ScreenSpot Pro.

How many models are evaluated on ScreenSpot Pro?

12 AI models have been evaluated on ScreenSpot Pro on BenchLM.

Last updated: April 10, 2026 · BenchLM version ScreenSpot Pro 2025

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.