Initialization of WebsiteSearchTool Without OpenAI Key — Embedder Provider Question

CrewAI Community Support
1

Hi,

I’m trying to configure the WebsiteSearchTool without relying on an OpenAI API key. I followed the example from the docs here:

tool = WebsiteSearchTool(
    config=dict(
        llm=dict(
            provider="ollama",  # or google, openai, anthropic, llama2, ...
            config=dict(
                model="llama2",
                # temperature=0.5,
                # top_p=1,
                # stream=true,
            ),
        ),
        embedder=dict(
            provider="google",  # or openai, ollama, ...
            config=dict(
                model="models/embedding-001",
                task_type="retrieval_document",
                # title="Embeddings",
            ),
        ),
    )
)

I modified it to this:

tool = WebsiteSearchTool(
    config=dict(
        llm=dict(
            provider="ollama",
            config=dict(
                model="deepseek-r1:14b",
            ),
        ),
        embedder=dict(
            provider="ollama",
            config=dict(
                model="models/embedding-001",
                task_type="retrieval_document",
            ),
        ),
    )
)

However, I’m unsure about the embedder provider configuration. Specifically, I want to use the nomic-embed-text model for embeddings.

My questions are:

  • How can I configure the WebsiteSearchTool to use nomic-embed-text as the embedder provider?
  • Is there a recommended way to integrate custom embedding models (like nomic-embed-text) into this framework?

Thanks in advance for any guidance!