Skip to content

OpenAI Context

The OpenaiContext is our default configuration for LaVague. It uses OpenAI's gpt-4o by default as LaVague's multi-modal mm_llm and ActionEngine LLM models, and OpenAI's text-embedding-3-small model as the embedding model. You can swap these models to other OpenAI models as we explain below.

Pre-requisites

The OpenaiContext is installed by default with LaVague.

API keys

You will need to set your OpenAI API keys as an OPENAI_API_KEY environment variable.

You can also pass your API keys to the OpenaiContext using the api_key parameters if initializing an OpenaiContext object directly.

For guidance on setting environment variables in your environment, see our FAQ.

Default usage

Since the OpenaiContext is used by default, there is no need to initialize or pass this context unless you wish to customize it.

Customizing the OpenaiContext

You can change the Openai models used by the context by initializing an OpenaiContext object and passing the name of the model(s) you wish to use to the llm, mm_llm and/or embedding parameters respectively.

You can find a list of OpenAI models currently available and their names here.

from lavague.contexts.openai import OpenaiContext

# Initialize Context
context = OpenaiContext(llm="gpt-4o-mini", mm_llm="gpt-4o-mini")
Once we have defined our OpenaiContext, we can use it by passing it to our ActionEngine and WorldModel when initializing them with the from_context() method.

selenium_driver = SeleniumDriver()

# Build Action Engine and World Model from Context
action_engine = ActionEngine.from_context(context=context, driver=selenium_driver)
world_model = WorldModel.from_context(context)

# Build agent & run query
agent = WebAgent(world_model, action_engine)
agent.get("https://huggingface.co/")
agent.run("What is this week's top Space of the week?")