Update app.py

app.py

@@ -368,6 +368,72 @@ For more information on `huggingface_hub` Inference API support, please check th
 
 # if __name__ == "__main__":
 #     demo.launch()
+# import os
+# import gradio as gr
+# from huggingface_hub import InferenceClient
+# from dotenv import load_dotenv
+
+# # Load environment variables
+# load_dotenv()
+# HF_TOKEN = os.getenv("HF_TOKEN")  # Ensure this is set in .env
+
+# # 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):
+#     max_tokens = 2048
+#     temperature = 0.7
+#     top_p = 0.95
+
+#     # Convert chat history into OpenAI-style format
+#     messages = [{"role": "system", "content": system_message}]
+#     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 = ""
+
+#     # 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,
+#         max_tokens=max_tokens,
+#         stream=True,
+#         temperature=temperature,
+#         top_p=top_p,
+#     ):
+#         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()# app.py
+
 import os
 import gradio as gr
 from huggingface_hub import InferenceClient
@@ -375,15 +441,15 @@ from dotenv import load_dotenv
 
 # Load environment variables
 load_dotenv()
-HF_TOKEN = os.getenv("HF_TOKEN")  # Ensure this is set in .env
+HF_TOKEN = os.getenv("HF_TOKEN")
 
 # Initialize Hugging Face Inference Client
 client = InferenceClient(
-    model="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
+    model="mistralai/Devstral-Small-2505",
     token=HF_TOKEN
 )
 
-# Define system instructions for the chatbot
+# System prompt for coding assistant
 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. "
@@ -391,46 +457,34 @@ system_message = (
     "based on their requirements."
 )
 
-# Define the response generation function
+# Streaming chatbot logic
 def respond(message, history):
-    max_tokens = 2048
-    temperature = 0.7
-    top_p = 0.95
-
-    # Convert chat history into OpenAI-style format
+    # Prepare messages with system prompt
     messages = [{"role": "system", "content": system_message}]
-    for item in history:
-        role = item["role"]
-        content = item["content"]
-        messages.append({"role": role, "content": content})
-    
-    # Add the latest user message
+    for msg in history:
+        messages.append(msg)
     messages.append({"role": "user", "content": message})
 
+    # Stream response from the model
     response = ""
-
-    # Streaming response from the Hugging Face Inference API
     for chunk in client.chat.completions.create(
-        model="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
+        model="mistralai/Devstral-Small-2505",
         messages=messages,
-        max_tokens=max_tokens,
+        max_tokens=1024,
+        temperature=0.7,
+        top_p=0.95,
         stream=True,
-        temperature=temperature,
-        top_p=top_p,
     ):
-        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
-)
+        token = chunk.choices[0].delta.get("content", "") or ""
+        response += token
+        yield response
 
+# Create Gradio interface
+with gr.Blocks() as demo:
+    chatbot = gr.Chatbot(type='messages')  # Use modern message format
+    gr.ChatInterface(fn=respond, chatbot=chatbot, type="messages")  # Match format
+
+# Launch app
 if __name__ == "__main__":
     demo.launch()
 
@@ -441,4 +495,3 @@ if __name__ == "__main__":
 
 
 
-