multimodal benchmark
OSWorld-G
OSWorld-G (Grounding) evaluates screenshot grounding accuracy for OS automation tasks.
Updated Aug 7, 2026
Published models1
Registry coverage1
MetricScore
EvidenceB
OSWorld-G leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 0.683 points | 100.0% | 1 | C |
OSWorld-G highlights
The top published results on this benchmark's own scale.
What is OSWorld-G?
Definition and scoring fields from the benchmark registry.
OSWorld-G (Grounding) evaluates screenshot grounding accuracy for OS automation tasks.
Scores are shown in points. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- OSWorld-G
- Modality
- image
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- osworld-g|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about OSWorld-G.
Which model scores highest on OSWorld-G?
Qwen3 VL 235B A22B Thinking is currently ranked first with 0.683 points.
What does OSWorld-G measure?
OSWorld-G (Grounding) evaluates screenshot grounding accuracy for OS automation tasks.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.