A Large Action Model framework for developing AI Web Agents.
What is LaVague?
LaVague is an open-source framework specifically designed for developers eager to create AI Web Agents that can automate various processes for users. Imagine your web agent as a personal assistant that can take complex tasks and break them down into manageable steps. For example, if you want to install the Hugging Face Diffusers library, your agent can generate the necessary instructions and execute them seamlessly.
How Does LaVague Work?
At the core of LaVague, there are two main components that facilitate this automation:
- World Model: This component takes your objective (what you want to accomplish) along with the current web page state, and then it generates appropriate instructions to proceed.
- Action Engine: This part compiles the instructions into actionable code, which can be executed using popular automation tools like Selenium or Playwright.
Getting Started with LaVague
Ready to dive in? Here’s a step-by-step guide to getting started:
Demo
Check out how LaVague can take multiple steps to achieve the objective of “Go on the quick tour of PEFT”:
Hands-on Implementation
You can get started with these easy steps:
- Download LaVague with:
pip install lavague - Use our framework to build a Web Agent. Implement your objective using the following code:
- Launch the Gradio Agent Demo by running:
agent.demo("Go on the quick tour of PEFT")
from lavague.core import WorldModel, ActionEngine
from lavague.core.agents import WebAgent
from lavague.drivers.selenium import SeleniumDriver
selenium_driver = SeleniumDriver(headless=False)
world_model = WorldModel()
action_engine = ActionEngine(selenium_driver)
agent = WebAgent(world_model, action_engine)
agent.get("https://huggingface.co/docs")
agent.run("Go on the quick tour of PEFT")For further information on this example and how to use LaVague, explore our quick-tour.
Note: These examples use the default OpenAI API configuration, so you will need to set the OPENAI_API_KEY variable in your local environment for them to function correctly.
Key Features
- Built-in Contexts (configurations)
- Customizable configuration
- A test runner for performance benchmarking
- A Token Counter for estimating token usage and costs
- Logging tools
- An optional, interactive Gradio interface
- Debugging tools
- A Chrome Extension
Troubleshooting Guide
If you encounter any issues while getting started with LaVague, you can:
- Visit our troubleshooting guide for information and fixes on common problems.
- Open a GitHub issue with details about your problem.
- Contact us in the #support channel on our Discord server.
For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.
Contributing
Your input is invaluable to us! If you wish to contribute to the LaVague project, please check the outlined processes using GitHub issues to avoid duplicate efforts.
Final Thoughts
At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.
Good luck and happy coding with LaVague!