Benchmark
ScreenSpot Benchmark
What is ScreenSpot?
ScreenSpot is the first realistic GUI grounding benchmark that encompasses mobile, desktop, and web environments. The dataset comprises over 1,200 instructions…
- Released
- 2024-01-17
- Modality
- multimodal
- Evaluates
- multimodal, spatial reasoning, grounding, vision
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 16
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/screenspot?top_n=500
ScreenSpot results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Nova 2 Lite | Amazon | 0.833 | 2025-12-02 | evidence |
| Nova 2 Omni | Amazon | 0.854 | 2025-12-02 | evidence |
| Nova 2 Pro | Amazon | 0.881 | 2025-12-02 | evidence |
| Qwen2.5 VL 32B Instruct | Qwen | 0.885 | 2025-02-28 | evidence |
| Qwen2.5 VL 72B Instruct | Qwen | 0.871 | 2025-01-26 | evidence |
| Qwen2.5 VL 7B Instruct | Qwen | 0.847 | 2025-01-26 | evidence |
| Qwen3 VL 235B A22B Instruct | Qwen | 0.954 | 2025-09-22 | evidence |
| Qwen3 VL 235B A22B Thinking | Qwen | 0.954 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Instruct | Qwen | 0.947 | 2025-09-22 | evidence |
| Qwen3 VL 30B A3B Thinking | Qwen | 0.947 | 2025-09-22 | evidence |
| Qwen3 VL 32B Instruct | Qwen | 0.958 | 2025-09-22 | evidence |
| Qwen3 VL 32B Thinking | Qwen | 0.957 | 2025-09-22 | evidence |
| Qwen3 VL 4B Instruct | Qwen | 0.94 | 2025-09-22 | evidence |
| Qwen3 VL 4B Thinking | Qwen | 0.929 | 2025-09-22 | evidence |
| Qwen3 VL 8B Instruct | Qwen | 0.944 | 2025-09-22 | evidence |
| Qwen3 VL 8B Thinking | Qwen | 0.936 | 2025-09-22 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.