LaVague Weekly Update #2
LaVague: A Large Action Model framework for building Web Agents

LaVague Weekly Update #2

Welcome to our weekly updates, where we announce the latest news on our open-source framework for Web Agents & share what's coming next and how you can get involved!

Laura Yie

Welcome to our second weekly update!

Here’s what we’ve been up to this week...

🚀 Adding new engines

We launched a new three-engine approach to improve performance and efficiency. The World Model can now decide if the next instruction should be handled by the:

  • 🚄 Action Engine: Generates and executes Selenium code to perform an action on a web page
  • 🐍 Python Engine: Generates and executes code for tasks that do not involve navigating or interacting with a web page, such as extracting information from a web page
  • 🕹️ Navigation Control: Performs frequently required navigation tasks without needing to make any extra LLM calls. So far, we cover scroll up, scroll down & wait

📑 Adding new use cases & examples

We've added new use cases and examples to our docs, including an end-to-end example where we show how to build an agent that can apply to jobs with a PNG of your resume using open-source frameworks!


🛤️ Coming soon:

  • 🌊 Join us at our Webinar next Thursday!
  • We'll be launching our interactive Gradio interface for Web Agents.
  • More documentation!
  • An Eval module for evaluating performance.

📢 Help wanted:

Hop onto our Discord to chat with us about LaVague and thanks for following our journey!

See you next time!