multimodal benchmark
OSWorld Leaderboard
OSWorld: The first-of-its-kind scalable, real computer environment for multimodal agents, supporting task setup, execution-based evaluation, and interactive learning across Ubuntu, Windows, and macOS with 369 computer tasks involving real web and desktop applications, OS file I/O, and multi-application workflows
Updated Aug 17, 2026
Models20
Model coverage20
MetricScore
EvidenceB
OSWorld Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelBY | Score78.8% | Percentile100.0% | Participants20 | EvidenceC | Evaluated |
| Rank02 | ModelBY | Score76.4% | Percentile94.7% | Participants20 | EvidenceC | Evaluated |
| Rank03 | ModelAN | Score72.7% | Percentile89.5% | Participants20 | EvidenceC | Evaluated |
| Rank04 | ModelAN | Score72.5% | Percentile84.2% | Participants20 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score66.7% | Percentile79.0% | Participants20 | EvidenceC | Evaluated |
| Rank06 | ModelAN | Score66.3% | Percentile73.7% | Participants20 | EvidenceC | Evaluated |
| Rank07 | ModelZA | Score62.3% | Percentile68.4% | Participants20 | EvidenceC | Evaluated |
| Rank08 | ModelAN | Score61.4% | Percentile63.2% | Participants20 | EvidenceC | Evaluated |
| Rank09 | ModelAN | Score50.7% | Percentile57.9% | Participants20 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score41.0% | Percentile52.6% | Participants20 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score38.1% | Percentile47.4% | Participants20 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score33.9% | Percentile42.1% | Participants20 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score33.9% | Percentile36.8% | Participants20 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score32.6% | Percentile31.6% | Participants20 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score31.4% | Percentile26.3% | Participants20 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score30.6% | Percentile21.1% | Participants20 | EvidenceC | Evaluated |
| Rank17 | ModelAC | Score30.3% | Percentile15.8% | Participants20 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score26.2% | Percentile10.5% | Participants20 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score8.8% | Percentile5.3% | Participants20 | EvidenceC | Evaluated |
| Rank20 | ModelAC | Score5.9% | Percentile0.0% | Participants20 | EvidenceC | Evaluated |
OSWorld Highlights
The leading models and scores on this benchmark.
OSWorld Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for OSWorld
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis osworld 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.
- 01BYSeed 2.1 ProByteDance
- Score
- 78.8%
Strengths
- Ranks #1 of 20 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures OSWorld, not total model capability
- 02BYSeed 2.1 TurboByteDance
- Score
- 76.4%
Strengths
- Ranks #2 of 20 compared models
- 95th percentile on this benchmark
- C evidence result
Considerations
- This result measures OSWorld, not total model capability
- 03ANClaude Opus 4.6Anthropic
- Score
- 72.7%
- Price
- $5.0 input / $25 output per 1M tokens
- Speed
- Up to 17 tok/s via Anthropic
Strengths
- Ranks #3 of 20 compared models
- 89th percentile on this benchmark
- C evidence result
Considerations
- This result measures OSWorld, not total model capability
- 04ANClaude Sonnet 4.6Anthropic
- Score
- 72.5%
- Price
- $3.0 input / $15 output per 1M tokens
- Speed
- Up to 15 tok/s via Anthropic
Strengths
- Ranks #4 of 20 compared models
- 84th percentile on this benchmark
- C evidence result
Considerations
- This result measures OSWorld, not total model capability
- 05ACQwen3 VL 235B A22B InstructAlibaba Cloud / Qwen Team
- Score
- 66.7%
Strengths
- Ranks #5 of 20 compared models
- 79th percentile on this benchmark
- C evidence result
Considerations
- This result measures OSWorld, not total model capability
Selection summary
Best AI Models for OSWorld
Seed 2.1 Pro currently leads OSWorld with 78.8%. 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 OSWorld?
What OSWorld measures and how its scores work.
OSWorld: The first-of-its-kind scalable, real computer environment for multimodal agents, supporting task setup, execution-based evaluation, and interactive learning across Ubuntu, Windows, and macOS with 369 computer tasks involving real web and desktop applications, OS file I/O, and multi-application workflows
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- OSWorld
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- osworld|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about OSWorld.
Which model scores highest on OSWorld?
Seed 2.1 Pro is currently ranked first with 78.8%.
What does OSWorld measure?
OSWorld: The first-of-its-kind scalable, real computer environment for multimodal agents, supporting task setup, execution-based evaluation, and interactive learning across Ubuntu, Windows, and macOS with 369 computer tasks involving real web and desktop applications, OS file I/O, and multi-application workflows
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
20 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.