PDFKnowledgeSource with Ollama embedder error

llm1 = LLM(model="ollama/deepseek-r1:7b",temperature=0.1)

pdf_source = PDFKnowledgeSource(
    file_paths=["ChatGPT for Robotics: Design Principles and Model Abilities - Adaptive Task Planning.pdf"],
    chunk_size=2000,
    chunk_overlap=2,
)

retriever_agent = Agent(
    role="Knowledge Retriever",
    goal="Retrieve the most relevant information for answering questions.",
    backstory="This agent is a knowledge retrieval assistant. It uses FAISS and a large language model to find the most relevant context for any query, based on academic papers and research databases.",
    knowledge_sources=[pdf_source],
    llm=llm1,
    verbose=True,
    embedder=dict(
        provider="ollama",
        config=dict(
            model= "nomic-embed-text",
            api_base="http://localhost:11434"
        )
    )
)

I got below error:

ERROR:root:Failed to get supported params: argument of type ‘NoneType’ is not iterable

Provider List: Providers | liteLLM

How to solve this issue?