OSU-NLP-Group/SeeAct
SeeAct is a generalist web agent system that uses large multimodal models like GPT-4V to autonomously execute tasks on any website.
Not currently ranked — collecting fresh signals.
star history
SeeAct is a system for generalist web agents that autonomously carry out tasks on any given website, with a focus on large multimodal models (LMMs) such as GPT-4V(ision). It provides a robust codebase supporting running web agent experiments and includes the Multimodal-Mind2Web benchmark for evaluation. The system enables agents to ground visual understanding with web interactions for executing user-specified tasks.
Frequently asked
- What is OSU-NLP-Group/SeeAct?
- SeeAct is a generalist web agent system that uses large multimodal models like GPT-4V to autonomously execute tasks on any website.
- Is SeeAct open source?
- Yes — OSU-NLP-Group/SeeAct is an open-source project tracked on heatdrop.
- What language is SeeAct written in?
- OSU-NLP-Group/SeeAct is primarily written in Python.
- How popular is SeeAct?
- OSU-NLP-Group/SeeAct has 851 stars on GitHub.
- Where can I find SeeAct?
- OSU-NLP-Group/SeeAct is on GitHub at https://github.com/OSU-NLP-Group/SeeAct.