Chang to AzureOpenAI

app.py

@@ -5,6 +5,7 @@ from langchain.embeddings import HuggingFaceEmbeddings
 from langchain.llms import CTransformers
 from langchain.llms import GooglePalm
 from langchain.llms import Replicate
+from langchain.llms import AzureOpenAI
 from langchain.text_splitter import CharacterTextSplitter
 from langchain.vectorstores import FAISS
 from langchain.memory import ConversationBufferMemory
@@ -71,11 +72,16 @@ def create_conversational_chain(vector_store):
     #     model = "replicate/llama-2-70b-chat:58d078176e02c219e11eb4da5a02a7830a283b14cf8f94537af893ccff5ee781", 
     #     callbacks=[StreamingStdOutCallbackHandler()],
     #     input = {"temperature": 0.01, "max_length" :500,"top_p":1})
-    llm = GooglePalm(
+    # llm = GooglePalm(
+    #     streaming = True,
+    #     model = "models/text-bison-001", # "google/flan-t5-large"
+    #     callbacks=[StreamingStdOutCallbackHandler()],
+    #     input = {"temperature": 0.07, "max_length" :800,"top_p":0.95})
+    llm = AzureOpenAI(
         streaming = True,
-        model = "models/text-bison-001", #" "google/flan-t5-large"
-        callbacks=[StreamingStdOutCallbackHandler()],
-        input = {"temperature": 0.07, "max_length" :800,"top_p":0.95})
+        deployment_name="HCloudChat",
+        model_name="gpt-35-turbo",
+        callbacks=[StreamingStdOutCallbackHandler()])
     memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
 
     chain = ConversationalRetrievalChain.from_llm(llm=llm, chain_type='stuff',

requirements.txt

@@ -12,4 +12,5 @@ pypdf
 python-dotenv
 replicate
 docx2txt
-google-generativeai
+google-generativeai
+openai