multimodal benchmark
GroundUI-1K
A subset of GroundUI-18K for UI grounding evaluation, where models must predict action coordinates on screenshots based on single-step instructions across web, desktop, and mobile platforms.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
GroundUI-1K leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AM Nova ProAmazon | 81.4% | 100.0% | 2 | C | |
| 2 | AMNova LiteAmazon | 80.2% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
GroundUI-1K highlights
The top published results on this benchmark's own scale.
What is GroundUI-1K?
Definition and scoring fields from the benchmark registry.
A subset of GroundUI-18K for UI grounding evaluation, where models must predict action coordinates on screenshots based on single-step instructions across web, desktop, and mobile platforms.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- GroundUI-1K
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- groundui-1k|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about GroundUI-1K.
Which model scores highest on GroundUI-1K?
Nova Pro is currently ranked first with 81.4%.
What does GroundUI-1K measure?
A subset of GroundUI-18K for UI grounding evaluation, where models must predict action coordinates on screenshots based on single-step instructions across web, desktop, and mobile platforms.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 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.