LLM Reference
activeAgents

ScreenSpot-Pro

Metric: Accuracy (higher is better)Introduced: 2025

Harder ScreenSpot-family UI grounding benchmark for locating interface elements across visual computer-use tasks. High benchmark score alone doesn't make a model the right pick — weigh it against pricing, API availability, and release date.

Models ranked

5

tracked on this benchmark

Score band

78.5 – 51.5

best → lowest tracked

Snapshot trend

-1.47

Sep 25 → Apr 28 · 1 models

Leaderboard

Tracked models ranked by Accuracy (higher is better).

Compare candidates
#Model variant and provenanceScore
1
Holo2-235B-A22B
Version: ScreenSpot-Pro - agentic localization (3 steps)Harness: Not recordedEvaluator: Not recordedObserved: Nov 13, 2025Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes - keep separate from single-step variant.

78.5
2
Holo2-30B-A3B
Version: ScreenSpot-Pro - UI localizationHarness: Not recordedEvaluator: Not recordedObserved: Nov 13, 2025Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes.

66.1
3
Holo1.5-72B
Version: ScreenSpot-Pro - UI localizationHarness: Not recordedEvaluator: Not recordedObserved: Sep 25, 2025Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes.

63.3
4
Holotron-3-Nano
Version: ScreenSpot-Pro - UI localizationHarness: Not recordedEvaluator: Not recordedObserved: Apr 28, 2026Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes.

55.9
5
Holo1.5-3B
Version: ScreenSpot-Pro - UI localizationHarness: Not recordedEvaluator: Not recordedObserved: Sep 25, 2025Confidence: Not recordedSource

Notes: DAT-5701: H Company (self-reported); medium confidence; yes.

51.5

How to read this benchmark

This benchmark scores models where higher is better. Scores are useful for directional filtering and shortlisting — not for universal quality ranking. Prefer benchmarks closest to your workload, then validate the linked model pages for pricing, context window, and provider availability.

Trust this score when

  • There is a fresh timestamped snapshot (or multiple snapshots) for this benchmark.
  • The model list covers the same version family you can actually deploy today.
  • Top candidates overlap with your required routing and feature requirements.

Be cautious when

  • There is only one benchmark snapshot or the dataset appears stale.
  • The benchmark metric direction is opposite of your decision objective.
  • The score difference between options is narrow and likely within implementation variance.

FAQ

What does the ScreenSpot-Pro benchmark measure?

Harder ScreenSpot-family UI grounding benchmark for locating interface elements across visual computer-use tasks. On this page it lists 5 tracked model variants where higher is better.

Is a higher ScreenSpot-Pro score always better?

For this benchmark, higher is better. A high score helps you shortlist, but confirm pricing, context window, and provider availability on each model page before committing — the top scorer is not always the right pick for your workload or budget.

How current is this ScreenSpot-Pro data?

This benchmark was last reviewed on Jun 7, 2026. The tracked score average moved -1.47 points across the last 3 snapshots.

Related benchmarks

Last reviewed: Jun 7, 2026

Resources