I got llama index working with a

llm = AzureOpenAI(temperature=0, model_name=args.model_name, deployment_name=args.model_name)

index = GPTVectorStoreIndex.from_documents(documents, service_context=service_context)

openai.error.InvalidRequestError: Must provide an 'engine' or 'deployment_id' parameter to create a <class 'openai.api_resources.embedding.Embedding'>

index = GPTVectorStoreIndex.from_documents(documents, service_context=service_context)

def create_service_context():
    # define LLM
    if openai.api_type == "azure":
        # Use Azure OpenAI API
        logging.info("Using Azure OpenAI.")
        #llm = AzureOpenAI(temperature=0, model_name=args.model_name, deployment_name=args.model_name)
        llm = AzureOpenAI(
            deployment_name=args.model_name,
            temperature=0,
            model_kwargs={
                "api_key": openai.api_key,
                "api_base": openai.api_base,
                "api_type": openai.api_type,
                "api_version": openai.api_version,
            }
        )
    else:
        # Using OpenAI API
        if args.model_name == "gpt-3.5-turbo":
            logging.info("Using ChatOpenAI.")
            llm = ChatOpenAI(temperature=0, model_name="gpt-3.5-turbo")
        else:
            logging.info("Using OpenAI.")
            llm = OpenAI(temperature=0, model_name=args.model_name)

    # Define LLMPredictor
    llm_predictor = LLMPredictor(llm=llm)

    # Define Embedding Model
    embedding_llm = LangchainEmbedding(
        OpenAIEmbeddings(
            model="text-embedding-ada-002",
            deployment="text-embedding-ada-002",
            openai_api_key=openai.api_key,
            openai_api_base=openai.api_base,
            openai_api_type=openai.api_type,
            openai_api_version=openai.api_version,
        ),
        embed_batch_size=1,
    )

    # define prompt helper
    # set maximum input size
    max_input_size = 4096

    # set number of output tokens
    num_output = 256
    
    # set maximum chunk overlap
    max_chunk_overlap = 20
    
    prompt_helper = PromptHelper(max_input_size, num_output, max_chunk_overlap)

    return ServiceContext.from_defaults(
        llm_predictor=llm_predictor,
        prompt_helper=prompt_helper,
        embed_model=embedding_llm
    )

from llama_index import ServiceContext, GPTVectorStoreIndex, LLMPredictor, LangchainEmbedding, SimpleDirectoryReader
from langchain.chat_models import AzureChatOpenAI
from langchain.embeddings import OpenAIEmbeddings

API_KEY = "<my key>"
API_BASE = "https://<resource name>.openai.azure.com/"
files_path = "/Users/abissell/az/files/"
embed_deployment_name = ""
chat_deployment_name = ""

embedding = OpenAIEmbeddings(
    deployment=embed_deployment_name,
    openai_api_key=API_KEY,
    openai_api_base=API_BASE,
    openai_api_type="azure",
    openai_api_version="2022-12-01",
)
embedding_llm = LangchainEmbedding(embedding, embed_batch_size=1)

llm = AzureChatOpenAI(
    deployment_name=chat_deployment_name,
    openai_api_key=API_KEY,
    openai_api_base=API_BASE,
    openai_api_type="azure",
    openai_api_version="2023-03-15-preview",
)
llm_predictor = LLMPredictor(llm=llm)

service_context = ServiceContext.from_defaults(
    llm_predictor=llm_predictor,
    embed_model=embedding_llm,
)
documents = SimpleDirectoryReader(files_path).load_data()
index = GPTVectorStoreIndex.from_documents(documents, service_context=service_context)
query_engine = index.as_query_engine()
print(query_engine.query("sample"))