Update app.py

app.py

@@ -18,6 +18,7 @@ from langchain_community.document_loaders.blob_loaders.youtube_audio import Yout
 from langchain.schema import AIMessage, HumanMessage
 from langchain_community.llms import HuggingFaceHub
 from langchain_huggingface import HuggingFaceEndpoint
+from langchain_community.llms import HuggingFaceEndPoints
 from langchain_huggingface import HuggingFaceEmbeddings
 from langchain_community.llms import HuggingFaceTextGenInference
 #from langchain_community.embeddings import HuggingFaceInstructEmbeddings, HuggingFaceEmbeddings, HuggingFaceBgeEmbeddings, HuggingFaceInferenceAPIEmbeddings
@@ -212,7 +213,7 @@ def generate_text (prompt, chatbot, history, vektordatenbank, retriever, top_p=0
         #pipe = pipeline("text-generation", model=MODEL_NAME_HF, config={"temperature": 0.5, "max_length": 512, "num_return_sequences": 1, "top_k": top_k, "top_p": top_p, "repetition_penalty": repetition_penalty})
 
         # Erstelle eine HuggingFaceEndPoints-Instanz mit den entsprechenden Endpunkt-Parametern
-        llm = HuggingFaceEndPoints(
+        llm = HuggingFaceEndPoint(
             endpoint_url=f"https://api-inference.huggingface.co/models/{MODEL_NAME_HF}",
             api_key=hf_token,
             model_kwargs=model_kwargs