Add Inference Client

app.py

@@ -7,15 +7,16 @@ from langchain_community.llms.huggingface_pipeline import HuggingFacePipeline
 from langchain_community.llms import HuggingFaceEndpoint
 from langchain.prompts import PromptTemplate, ChatPromptTemplate
 from langchain_core.prompts.chat import SystemMessagePromptTemplate, HumanMessagePromptTemplate
+from huggingface_hub import InferenceClient
 import os
 import gradio as gr
 import spaces
 
-your_endpoint_url = "https://kp4xdy196cw81uf3.us-east-1.aws.endpoints.huggingface.cloud"
+endpoint_url = "https://kp4xdy196cw81uf3.us-east-1.aws.endpoints.huggingface.cloud"
 token = os.environ["API_TOKEN"]
 
 llm = HuggingFaceEndpoint(
-    endpoint_url=f"{your_endpoint_url}",
+    endpoint_url=f"{endpoint_url}",
     huggingfacehub_api_token = f"{token}",
     task = "text-generation",
     max_new_tokens=128,
@@ -27,6 +28,8 @@ llm = HuggingFaceEndpoint(
 )
 #print(llm)
 
+client = InferenceClient(endpoint_url, token=token)
+
 def chat_template_prompt():
     template = """
     Do not repeat questions and do not generate answer for user/human.Do not repeat yourself and do not create/generate dialogues.