multimodal benchmark
OSWorld-Verified Leaderboard
OSWorld-Verified is a verified subset of OSWorld, a scalable real computer environment for multimodal agents across Ubuntu, Windows, and macOS.
Updated Sep 5, 2026
Models24
Model coverage24
MetricScore
EvidenceB
OSWorld-Verified Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score86.10% | Percentile100.00% | Participants24 | EvidenceC | Evaluated |
| Rank02 | ModelAN | Score85.00% | Percentile95.65% | Participants24 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score84.30% | Percentile91.30% | Participants24 | EvidenceC | Evaluated |
| Rank04 | ModelAN | Score83.40% | Percentile86.96% | Participants24 | EvidenceC | Evaluated |
| Rank05 | ModelGO | Score83.00% | Percentile82.61% | Participants24 | EvidenceC | Evaluated |
| Rank06 | ModelAN | Score81.20% | Percentile78.26% | Participants24 | EvidenceC | Evaluated |
| Rank07 | ModelME | Score80.80% | Percentile73.91% | Participants24 | EvidenceC | Evaluated |
| Rank08 | ModelAN | Score79.60% | Percentile69.57% | Participants24 | EvidenceC | Evaluated |
| Rank09 | ModelOP | Score78.70% | Percentile65.22% | Participants24 | EvidenceC | Evaluated |
| Rank10 | ModelGO | Score78.40% | Percentile60.87% | Participants24 | EvidenceC | Evaluated |
| Rank11 | ModelAN | Score78.00% | Percentile56.52% | Participants24 | EvidenceC | Evaluated |
| Rank12 | ModelOP | Score75.00% | Percentile52.17% | Participants24 | EvidenceC | Evaluated |
| Rank13 | ModelGO | Score74.00% | Percentile47.83% | Participants24 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score73.30% | Percentile43.48% | Participants24 | EvidenceC | Evaluated |
| Rank15 | ModelMA | Score73.10% | Percentile39.13% | Participants24 | EvidenceC | Evaluated |
| Rank16 | ModelOP | Score72.10% | Percentile34.78% | Participants24 | EvidenceC | Evaluated |
| Rank17 | ModelMI | Score70.06% | Percentile30.43% | Participants24 | EvidenceC | Evaluated |
| Rank18 | ModelME | Score65.90% | Percentile26.09% | Participants24 | EvidenceC | Evaluated |
| Rank19 | ModelOP | Score64.70% | Percentile21.74% | Participants24 | EvidenceC | Evaluated |
| Rank20 | ModelAC | Score62.50% | Percentile17.39% | Participants24 | EvidenceC | Evaluated |
| Rank21 | ModelAC | Score58.00% | Percentile13.04% | Participants24 | EvidenceC | Evaluated |
| Rank22 | ModelAC | Score56.20% | Percentile8.70% | Participants24 | EvidenceC | Evaluated |
| Rank23 | ModelAC | Score54.50% | Percentile4.35% | Participants24 | EvidenceC | Evaluated |
| Rank24 | ModelOP | Score39.00% | Percentile0.00% | Participants24 | EvidenceC | Evaluated |
OSWorld-Verified Highlights
The leading models and scores on this benchmark.
OSWorld-Verified Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for OSWorld-Verified
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is OSWorld-Verified?
What OSWorld-Verified measures and how its scores work.
OSWorld-Verified is a verified subset of OSWorld that supports task setup, execution-based evaluation, and interactive learning for multimodal agents.
It uses the Score metric, expressed as a ratio, for execution-based evaluation.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- OSWorld-Verified
- 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 OSWorld-Verified.
Which model scores highest on OSWorld-Verified?
Qwen3.8 Max is currently ranked first with 86.10%.
What are the top three models on OSWorld-Verified?
The current leaders are Qwen3.8 Max (86.10%), Claude Fable 5 (85.00%), and Qwen3.8-27B (84.30%).
Which OSWorld-Verified model has the lowest official input price?
GPT-5.4 nano has the lowest matched official input price at $0.20 input / $1.3 output per 1M tokens.
Which models are fastest among OSWorld-Verified results?
The fastest matched records are Gemini 3.5 Flash (210.94 tok/s via Google), GPT-5.5 (134.94 tok/s via OpenAI), and GPT-5.4 nano (118.31 tok/s via OpenAI).
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.