[Python] TypeError: 'NoneType' object is not iterable when using ChatHuggingFace with...

I'm trying to use the TinyLlama/TinyLlama-1.1B-Chat-v1.0 model from Hugging Face with LangChain using the langchain_huggingface integration. My goal is to get a simple response from the model using ChatHuggingFace. Here's the code:

from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
from dotenv import load_dotenv

load_dotenv()

llm = HuggingFaceEndpoint(
repo_id="TinyLlama/TinyLlama-1.1B-Chat-v1.0",
task="text-generation"
)

model = ChatHuggingFace(llm=llm)

result = model.invoke(input="What is the capital of India")

print(result)



But when I run it, I get this error:

File "...huggingface.py", line 527, in _create_chat_result
for res in response["choices"]:
TypeError: 'NoneType' object is not iterable



It seems like response is None, which causes this error when trying to loop through response["choices"].

• 
  Verified my Hugging Face token is set via .env and works.
  
  
• 
  Model loads successfully on the Hugging Face web UI.
  
  
• 
  Used "text-generation" as the task based on the model card.
  
  
• 
  Expected model.invoke(...) to return a valid response, but got a NoneType error.
  
  
• 
  Not sure if the model is compatible with ChatHuggingFace or if a different task/prompt format is needed.
  

Continue reading...