multimodal benchmark
ScreenSpot Pro Leaderboard
ScreenSpot-Pro is a novel GUI grounding benchmark designed to rigorously evaluate the grounding capabilities of multimodal large language models (MLLMs) in professional high-resolution computing environments. The benchmark comprises 1,581 instructions across 23 applications spanning 5 industries and 3 operating systems, featuring authentic high-resolution images from professional domains with expert annotations. Unlike previous benchmarks that focus on cropped screenshots in consumer applications, ScreenSpot-Pro addresses the complexity and diversity of real-world professional software scenarios, revealing significant performance gaps in current MLLM GUI perception capabilities.
Updated Aug 17, 2026
Models25
Model coverage25
MetricScore
EvidenceC
ScreenSpot Pro Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Score87.9% | Percentile100.0% | Participants25 | EvidenceC | Evaluated |
| Rank02 | ModelOP | Score86.3% | Percentile95.8% | Participants25 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score84.5% | Percentile91.7% | Participants25 | EvidenceC | Evaluated |
| Rank04 | ModelME | Score84.1% | Percentile87.5% | Participants25 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score79.0% | Percentile83.3% | Participants25 | EvidenceC | Evaluated |
| Rank06 | ModelME | Score75.4% | Percentile79.2% | Participants25 | EvidenceC | Evaluated |
| Rank07 | ModelGO | Score72.7% | Percentile75.0% | Participants25 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score70.4% | Percentile70.8% | Participants25 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score70.3% | Percentile66.7% | Participants25 | EvidenceC | Evaluated |
| Rank10 | ModelGO | Score69.1% | Percentile62.5% | Participants25 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score68.6% | Percentile58.3% | Participants25 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score68.2% | Percentile54.2% | Participants25 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score62.0% | Percentile50.0% | Participants25 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score61.8% | Percentile45.8% | Participants25 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score60.5% | Percentile41.7% | Participants25 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score59.5% | Percentile37.5% | Participants25 | EvidenceC | Evaluated |
| Rank17 | ModelAC | Score57.9% | Percentile33.3% | Participants25 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score57.3% | Percentile29.2% | Participants25 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score57.1% | Percentile25.0% | Participants25 | EvidenceC | Evaluated |
| Rank20 | ModelAC | Score54.6% | Percentile20.8% | Participants25 | EvidenceC | Evaluated |
| Rank21 | ModelAC | Score49.2% | Percentile16.7% | Participants25 | EvidenceC | Evaluated |
| Rank22 | ModelAC | Score46.6% | Percentile12.5% | Participants25 | EvidenceC | Evaluated |
| Rank23 | ModelAC | Score43.6% | Percentile8.3% | Participants25 | EvidenceC | Evaluated |
| Rank24 | ModelAC | Score39.4% | Percentile4.2% | Participants25 | EvidenceC | Evaluated |
| Rank25 | ModelAC | Score29.0% | Percentile0.0% | Participants25 | EvidenceC | Evaluated |
ScreenSpot Pro Highlights
The leading models and scores on this benchmark.
ScreenSpot Pro Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for ScreenSpot Pro
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis screenspot pro AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ANClaude Opus 4.8Anthropic
- Score
- 87.9%
- Price
- $5.0 input / $25 output per 1M tokens
- Speed
- Up to 46 tok/s via Anthropic
Strengths
- Ranks #1 of 25 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures ScreenSpot Pro, not total model capability
- 02OPGPT-5.2OpenAI
- Score
- 86.3%
- Price
- $1.8 input / $14 output per 1M tokens
- Speed
- Up to 100 tok/s via OpenAI
Strengths
- Ranks #2 of 25 compared models
- 96th percentile on this benchmark
- C evidence result
Considerations
- This result measures ScreenSpot Pro, not total model capability
- 03ACQwen3.8 MaxAlibaba Cloud / Qwen Team
- Score
- 84.5%
- Price
- $2.0 input / $6.0 output per 1M tokens
Strengths
- Ranks #3 of 25 compared models
- 92th percentile on this benchmark
- C evidence result
Considerations
- This result measures ScreenSpot Pro, not total model capability
- 04MEMuse SparkMeta
- Score
- 84.1%
Strengths
- Ranks #4 of 25 compared models
- 88th percentile on this benchmark
- C evidence result
Considerations
- This result measures ScreenSpot Pro, not total model capability
- 05ACQwen3.7-PlusAlibaba Cloud / Qwen Team
- Score
- 79.0%
- Price
- $0.50 input / $3.0 output per 1M tokens
Strengths
- Ranks #5 of 25 compared models
- 83th percentile on this benchmark
- C evidence result
Considerations
- This result measures ScreenSpot Pro, not total model capability
Selection summary
Best AI Models for ScreenSpot Pro
Claude Opus 4.8 currently leads ScreenSpot Pro with 87.9%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is ScreenSpot Pro?
What ScreenSpot Pro measures and how its scores work.
ScreenSpot-Pro is a novel GUI grounding benchmark designed to rigorously evaluate the grounding capabilities of multimodal large language models (MLLMs) in professional high-resolution computing environments. The benchmark comprises 1,581 instructions across 23 applications spanning 5 industries and 3 operating systems, featuring authentic high-resolution images from professional domains with expert annotations. Unlike previous benchmarks that focus on cropped screenshots in consumer applications, ScreenSpot-Pro addresses the complexity and diversity of real-world professional software scenarios, revealing significant performance gaps in current MLLM GUI perception capabilities.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of C.
- Family
- ScreenSpot Pro
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- screenspot-pro|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about ScreenSpot Pro.
Which model scores highest on ScreenSpot Pro?
Claude Opus 4.8 is currently ranked first with 87.9%.
What does ScreenSpot Pro measure?
ScreenSpot-Pro is a novel GUI grounding benchmark designed to rigorously evaluate the grounding capabilities of multimodal large language models (MLLMs) in professional high-resolution computing environments. The benchmark comprises 1,581 instructions across 23 applications spanning 5 industries and 3 operating systems, featuring authentic high-resolution images from professional domains with expert annotations. Unlike previous benchmarks that focus on cropped screenshots in consumer applications, ScreenSpot-Pro addresses the complexity and diversity of real-world professional software scenarios, revealing significant performance gaps in current MLLM GUI perception capabilities.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
25 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.