Update app.py

app.py

@@ -1,64 +1,70 @@
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
+# Initialize model and tokenizer
+model = AutoModelForCausalLM.from_pretrained(
+    "microsoft/phi-3-mini-instruct",
+    trust_remote_code=True,
+    torch_dtype=torch.float32,  # Force float32 for CPU compatibility
+    device_map="auto",          # Automatically handles CPU-only environment
+)
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+tokenizer = AutoTokenizer.from_pretrained("microsoft/phi-3-mini-instruct")
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Create pipeline
+pipe = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    device=-1,  # Force CPU usage (-1 = CPU)
+)
 
+def respond(message, chat_history):
+    # Format the chat history with system prompt
+    system_prompt = "You are a helpful AI assistant."
+    messages = [{"role": "system", "content": system_prompt}]
+    
+    # Add previous chat history
+    for user_msg, bot_msg in chat_history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": bot_msg})
+    
+    # Add new user message
     messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
+    
+    # Format for the model
+    formatted_input = tokenizer.apply_chat_template(
         messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
+        tokenize=False,
+        add_generation_prompt=True
+    )
+    
+    # Generate response
+    generation_args = {
+        "max_new_tokens": 200,  # Reduced for CPU performance
+        "temperature": 0.0,
+        "do_sample": False,
+    }
+    
+    output = pipe(formatted_input, **generation_args)
+    response = output[0]['generated_text']
+    
+    # Remove the input text from the response
+    if response.startswith(formatted_input):
+        response = response[len(formatted_input):]
+    
+    return response
 
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Create Gradio interface
 demo = gr.ChatInterface(
     respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+    chatbot=gr.Chatbot(height=400),
+    textbox=gr.Textbox(placeholder="Ask me anything...", container=False, scale=7),
+    title="Phi-3 Mini Chat (CPU)",
+    examples=["What's 2x + 3 = 7?", "How to make banana smoothie?", "Explain quantum computing simply"],
+    cache_examples=False,
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()