Update app.py

app.py

@@ -309,6 +309,63 @@ For more information on `huggingface_hub` Inference API support, please check th
 
 # demo = gr.ChatInterface(respond, type="messages")
 
+# if __name__ == "__main__":
+#     demo.launch()
+# import os
+# import gradio as gr
+# from huggingface_hub import InferenceClient
+# from dotenv import load_dotenv
+
+# # Load .env variables (make sure to have HF_TOKEN in .env or set as env var)
+# load_dotenv()
+# HF_TOKEN = os.getenv("HF_TOKEN")  # or directly assign your token here as string
+
+# # Initialize InferenceClient with Hugging Face API token
+# client = InferenceClient(
+#     model="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
+#     token=HF_TOKEN
+# )
+
+# def respond(message, history):
+#     """
+#     Chat response generator function streaming from Hugging Face Inference API.
+#     """
+#     system_message = (
+#         "You are a helpful and experienced coding assistant specialized in web development. "
+#         "Help the user by generating complete and functional code for building websites. "
+#         "You can provide HTML, CSS, JavaScript, and backend code (like Flask, Node.js, etc.) "
+#         "based on their requirements."
+#     )
+#     max_tokens = 2048
+#     temperature = 0.7
+#     top_p = 0.95
+
+#     # Prepare messages in OpenAI chat format
+#     messages = [{"role": "system", "content": system_message}]
+#     for user_msg, assistant_msg in history:
+#         if user_msg:
+#             messages.append({"role": "user", "content": user_msg})
+#         if assistant_msg:
+#             messages.append({"role": "assistant", "content": assistant_msg})
+#     messages.append({"role": "user", "content": message})
+
+#     response = ""
+#     # Stream response tokens from Hugging Face Inference API
+#     for chunk in client.chat.completions.create(
+#         model="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
+#         messages=messages,
+#         max_tokens=max_tokens,
+#         stream=True,
+#         temperature=temperature,
+#         top_p=top_p,
+#     ):
+#         token = chunk.choices[0].delta.get("content", "")
+#         response += token
+#         yield response
+
+# # Create Gradio chat interface
+# demo = gr.ChatInterface(fn=respond, title="Website Building Assistant")
+
 # if __name__ == "__main__":
 #     demo.launch()
 import os
@@ -316,41 +373,43 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 from dotenv import load_dotenv
 
-# Load .env variables (make sure to have HF_TOKEN in .env or set as env var)
+# Load environment variables
 load_dotenv()
-HF_TOKEN = os.getenv("HF_TOKEN")  # or directly assign your token here as string
+HF_TOKEN = os.getenv("HF_TOKEN")  # Ensure this is set in .env
 
-# Initialize InferenceClient with Hugging Face API token
+# Initialize Hugging Face Inference Client
 client = InferenceClient(
     model="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
     token=HF_TOKEN
 )
 
+# Define system instructions for the chatbot
+system_message = (
+    "You are a helpful and experienced coding assistant specialized in web development. "
+    "Help the user by generating complete and functional code for building websites. "
+    "You can provide HTML, CSS, JavaScript, and backend code (like Flask, Node.js, etc.) "
+    "based on their requirements."
+)
+
+# Define the response generation function
 def respond(message, history):
-    """
-    Chat response generator function streaming from Hugging Face Inference API.
-    """
-    system_message = (
-        "You are a helpful and experienced coding assistant specialized in web development. "
-        "Help the user by generating complete and functional code for building websites. "
-        "You can provide HTML, CSS, JavaScript, and backend code (like Flask, Node.js, etc.) "
-        "based on their requirements."
-    )
     max_tokens = 2048
     temperature = 0.7
     top_p = 0.95
 
-    # Prepare messages in OpenAI chat format
+    # Convert chat history into OpenAI-style format
     messages = [{"role": "system", "content": system_message}]
-    for user_msg, assistant_msg in history:
-        if user_msg:
-            messages.append({"role": "user", "content": user_msg})
-        if assistant_msg:
-            messages.append({"role": "assistant", "content": assistant_msg})
+    for item in history:
+        role = item["role"]
+        content = item["content"]
+        messages.append({"role": role, "content": content})
+    
+    # Add the latest user message
     messages.append({"role": "user", "content": message})
 
     response = ""
-    # Stream response tokens from Hugging Face Inference API
+
+    # Streaming response from the Hugging Face Inference API
     for chunk in client.chat.completions.create(
         model="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
         messages=messages,
@@ -359,12 +418,18 @@ def respond(message, history):
         temperature=temperature,
         top_p=top_p,
     ):
-        token = chunk.choices[0].delta.get("content", "")
-        response += token
-        yield response
-
-# Create Gradio chat interface
-demo = gr.ChatInterface(fn=respond, title="Website Building Assistant")
+        token = chunk.choices[0].delta.get("content")
+        if token is not None:
+            response += token
+            yield response
+
+# Create Gradio Chat Interface
+demo = gr.ChatInterface(
+    fn=respond,
+    title="Website Building Assistant",
+    chatbot=gr.Chatbot(show_label=False),
+    type="openai",  # Use OpenAI-style message format
+)
 
 if __name__ == "__main__":
     demo.launch()
@@ -376,3 +441,4 @@ if __name__ == "__main__":
 
 
 
+