Unable to specify http_client as a parameter to OpenAILLM

[yohei1126]

Hi,

We have a self hosted litellm gateway in front of Azure Open AI API and Gemini API. https://www.litellm.ai
A self signed private root CA certificate is used to sign a server certificate of this server. so I tried to private http_client with private root CA certificate as a parameter to OpenAIEmbeddings and OpenAILLM class.

main.py

from neo4j import GraphDatabase
from neo4j_graphrag.retrievers import VectorRetriever
from neo4j_graphrag.llm import OpenAILLM
from neo4j_graphrag.generation import GraphRAG
from neo4j_graphrag.embeddings import OpenAIEmbeddings
import os
import httpx

# 1. Neo4j driver
URI = "neo4j://localhost:7687"
AUTH = ("neo4j", "password123")
INDEX_NAME = "index-name"
AI_GW_BASE_URL = os.getenv("AI_GW_BASE_URL")
AI_GW_API_KEY = os.getenv("AI_GW_API_KEY")

# Path to your CA certificate
CA_CERT_PATH = "./Root-CA-V2.pem"

# Create HTTP client with custom CA certificate
http_client = httpx.Client(verify=CA_CERT_PATH)

# Connect to Neo4j database
driver = GraphDatabase.driver(URI, auth=AUTH)

# 2. Retriever
# Create Embedder object, needed to convert the user question (text) to a vector
embedder = OpenAIEmbeddings(
    base_url=AI_GW_BASE_URL,
    api_key=AI_GW_API_KEY,
    model="text-embedding-3-large",
    http_client=http_client)

# 3. LLM
llm = OpenAILLM(
    base_url=AI_GW_BASE_URL,
    api_key=AI_GW_API_KEY,
    model_name="gpt-4o",
    model_params={"temperature": 0},
    http_client=http_client)
llm.invoke("what is graph RAG")

# Initialize the retriever
retriever = VectorRetriever(
    driver=driver,
    index_name=INDEX_NAME,
    embedder=embedder)

# Initialize the RAG pipeline
rag = GraphRAG(retriever=retriever, llm=llm)

# Query the graph
QUERY_TEXT = "How do I do similarity search in Neo4j?"
response = rag.search(query_text=QUERY_TEXT, retriever_config={"top_k": 5})
print(response.answer)

pyproject.toml

[project]
name = "dde-graph-rag-builder"
version = "0.1.0"
description = "Add your description here"
readme = "README.md"
requires-python = ">=3.12"
dependencies = [
    "httpx>=0.28.1",
    "neo4j-graphrag[openai]==1.9.0",
]

but the main.py script as follows

uv run main.py
Traceback (most recent call last):
  File "/Users/yohei.onishi/work/dde_graph_rag_builder/main.py", line 35, in <module>
    llm = OpenAILLM(
          ^^^^^^^^^^
  File "/Users/yohei.onishi/work/dde_graph_rag_builder/.venv/lib/python3.12/site-packages/neo4j_graphrag/llm/openai_llm.py", line 377, in __init__
    self.async_client = self.openai.AsyncOpenAI(**kwargs)
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/yohei.onishi/work/dde_graph_rag_builder/.venv/lib/python3.12/site-packages/openai/_client.py", line 473, in __init__
    super().__init__(
  File "/Users/yohei.onishi/work/dde_graph_rag_builder/.venv/lib/python3.12/site-packages/openai/_base_client.py", line 1392, in __init__
    raise TypeError(
TypeError: Invalid `http_client` argument; Expected an instance of `httpx.AsyncClient` but got <class 'httpx.Client'>

As both sync client and async client are initialised with the same parameters so we can not specify http_client.
https://github.com/neo4j/neo4j-graphrag-python/blob/main/src/neo4j_graphrag/llm/openai_llm.py#L375