llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model Directory

Scoring & Data

Scoring & Data
1224 models729 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll Benchmarks
llmboard.aiCopyright 2026 llmboard.ai

multimodal benchmark

ScreenSpot Pro Leaderboard

ScreenSpot Pro is a multimodal GUI grounding benchmark for evaluating MLLM performance in professional high-resolution computing environments.

Updated Sep 5, 2026

Models26
Model coverage26
MetricScore
EvidenceC

On this page

  • Ranking
  • Highlights
  • Distribution
  • Top models
  • About
  • FAQ

ScreenSpot Pro Ranking

Higher score ranks better on this benchmark.

26 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelOPGPT-6 AstraOpenAIScore92.70%Percentile100.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank02ModelANClaude Opus 4.8AnthropicScore87.90%Percentile96.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank03ModelOPGPT-5.2OpenAIScore86.30%Percentile92.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank04ModelACQwen3.8 MaxAlibaba Cloud / Qwen TeamScore84.50%Percentile88.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank05ModelMEMuse SparkMetaScore84.10%Percentile84.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank06ModelACQwen3.7-PlusAlibaba Cloud / Qwen TeamScore79.00%Percentile80.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank07ModelMEMuse Glimmer-30BMetaScore75.40%Percentile76.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank08ModelGOGemini 3 ProGoogleScore72.70%Percentile72.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank09ModelACQwen3.5-122B-A10BAlibaba Cloud / Qwen TeamScore70.40%Percentile68.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank10ModelACQwen3.5-27BAlibaba Cloud / Qwen TeamScore70.30%Percentile64.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank11ModelGOGemini 3 FlashGoogleScore69.10%Percentile60.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank12ModelACQwen3.5-35B-A3BAlibaba Cloud / Qwen TeamScore68.60%Percentile56.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank13ModelACQwen3.6 PlusAlibaba Cloud / Qwen TeamScore68.20%Percentile52.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank14ModelACQwen3 VL 235B A22B InstructAlibaba Cloud / Qwen TeamScore62.00%Percentile48.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank15ModelACQwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen TeamScore61.80%Percentile44.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank16ModelACQwen3 VL 30B A3B InstructAlibaba Cloud / Qwen TeamScore60.50%Percentile40.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank17ModelACQwen3 VL 4B InstructAlibaba Cloud / Qwen TeamScore59.50%Percentile36.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank18ModelACQwen3 VL 32B InstructAlibaba Cloud / Qwen TeamScore57.90%Percentile32.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank19ModelACQwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen TeamScore57.30%Percentile28.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank20ModelACQwen3 VL 32B ThinkingAlibaba Cloud / Qwen TeamScore57.10%Percentile24.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank21ModelACQwen3 VL 8B InstructAlibaba Cloud / Qwen TeamScore54.60%Percentile20.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank22ModelACQwen3 VL 4B ThinkingAlibaba Cloud / Qwen TeamScore49.20%Percentile16.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank23ModelACQwen3 VL 8B ThinkingAlibaba Cloud / Qwen TeamScore46.60%Percentile12.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank24ModelACQwen2.5 VL 72B InstructAlibaba Cloud / Qwen TeamScore43.60%Percentile8.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank25ModelACQwen2.5 VL 32B InstructAlibaba Cloud / Qwen TeamScore39.40%Percentile4.00%Participants26EvidenceCEvaluatedSep 8, 2026
Rank26ModelACQwen2.5 VL 7B InstructAlibaba Cloud / Qwen TeamScore29.00%Percentile0.00%Participants26EvidenceCEvaluatedSep 8, 2026

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.

ScreenSpot Pro

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.

What is ScreenSpot Pro?

What ScreenSpot Pro measures and how its scores work.

ScreenSpot Pro comprises 1,581 instructions across 23 applications, 5 industries, and 3 operating systems, using high-resolution images from professional domains with expert annotations.

It measures the GUI grounding capabilities of multimodal large language models in professional software scenarios.

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
No
Evaluation key
overall

LLM Stats. 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?

GPT-6 Astra is currently ranked first with 92.70%.

What are the top three models on ScreenSpot Pro?

The current leaders are GPT-6 Astra (92.70%), Claude Opus 4.8 (87.90%), and GPT-5.2 (86.30%).

Which ScreenSpot Pro model has the lowest official input price?

Qwen3.5-35B-A3B has the lowest matched official input price at $0.25 input / $2.0 output per 1M tokens.

Which models are fastest among ScreenSpot Pro results?

The fastest matched records are Gemini 3 Flash (124.32 tok/s via Google), GPT-5.2 (100.00 tok/s via OpenAI), and Gemini 3 Pro (90.00 tok/s via Google).

Does the highest score result prove overall model quality?

No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.

What does ScreenSpot Pro measure?

It measures the GUI grounding capabilities of multimodal large language models in professional software scenarios.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

26 model results are currently shown.

Does this benchmark affect the overall score?

No. This benchmark is shown for reference but does not contribute to the overall score.

Rank #1GPT-6 Astra92.70%
Rank #2Claude Opus 4.887.90%
Rank #3GPT-5.286.30%
Rank #4Qwen3.8 Max84.50%

Ranking basisThis screenspot pro AI model leaderboard uses descending score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    OP
    GPT-6 AstraOpenAI
    Score
    92.70%
    Price
    $10 input / $50 output per 1M tokens
    Speed
    Up to 30.66 tok/s via OpenAI

    Strengths

    • Ranks #1 of 26 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures ScreenSpot Pro, not total model capability
  2. 02
    AN
    Claude Opus 4.8Anthropic
    Score
    87.90%
    Price
    $5.0 input / $25 output per 1M tokens
    Speed
    Up to 42.00 tok/s via Vertex AI

    Strengths

    • Ranks #2 of 26 compared models
    • 96th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures ScreenSpot Pro, not total model capability
  3. 03
    OP
    OpenAI
    Score
    86.30%
    Price
    $1.8 input / $14 output per 1M tokens
    Speed
    Up to 100.00 tok/s via OpenAI

    Strengths

    • Ranks #3 of 26 compared models
    • 92th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures ScreenSpot Pro, not total model capability
  4. 04
    AC
    Alibaba Cloud / Qwen Team
    Score
    84.50%
    Price
    $2.0 input / $6.0 output per 1M tokens
    Speed
    Up to 60.87 tok/s via DeepInfra

    Strengths

    • Ranks #4 of 26 compared models
    • 88th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures ScreenSpot Pro, not total model capability
  5. 05
    ME
    Meta
    Score
    84.10%

    Strengths

    • Ranks #5 of 26 compared models
    • 84th 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

GPT-6 Astra currently leads ScreenSpot Pro with 92.70%. 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.

GPT-5.2
Qwen3.8 Max
Muse Spark
Benchmark rank #1GPT-6 Astra92.70% · $10 input / $50 output per 1M tokens
Benchmark rank #2Claude Opus 4.887.90% · $5.0 input / $25 output per 1M tokens
Benchmark rank #3GPT-5.286.30% · $1.8 input / $14 output per 1M tokens