Tarsier
    
            
                
                    Vision utilities for web interaction agents
                
            
             
            
             
            
        
            ... as buttons, links, or input fields that are visible on the page; Tarsier can also tag all textual elements if you pass tag_text_elements=True. Furthermore, we've developed an OCR algorithm to convert a page screenshot into a whitespace-structured string (almost like ASCII art) that an LLM even without vision can understand. Since current vision-language models still lack fine-grained representations needed for web interaction tasks, this is critical.