Coding and agents
OSWorld-Verified
1 model with a published score, best first. Higher is better.
86.1%Qwen3.8-MaxRank 1, published Aug 3, 2026
Leaderboard
Each model's latest published score. Every score links to its source.
| # | Model | OSWorld-V | Date | Note | Source |
|---|---|---|---|---|---|
| 1 | A Qwen3.8-MaxRecordAlibaba | 86.1% | Aug 3, 2026 | Alibaba-reported | benchlm.ai |
Record history
The best score ever published on OSWorld-Verified, and who held it. Each step is a new record.
| Date | Record holder | OSWorld-V | Gain | Note |
|---|---|---|---|---|
| Aug 3, 2026 | Qwen3.8-MaxCurrent | 86.1% | First | Alibaba-reported |
Method
Higher is better. Board source: os-world.github.io.
We collect the scores the board and the labs publish and run none ourselves. Ranks use each model's latest reading. 1 reading stored for this board, the newest from Aug 3, 2026.