InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to Deliberative Reasoners
Open paper
Benchmark
Model
Rank
Results
natural-language-visual-grounding-on-screenspot
InfiGUI-R1-3B (extra data)
#12
Accuracy (%): 87.5
computer-use-agents-on-screenspot-pro
InfiGUI-R1-3B (extra data)
#43
Accuracy: 35.7
Rank counts only results with a code link.