Update app.py

app.py

@@ -1,33 +1,30 @@
 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForCausalLM
 from transformers.utils import get_json_schema
-import torch
 
 # -----------------------
-# Load model
+# Load model (CPU friendly)
 # -----------------------
 model_name = "bhaiyahnsingh45/functiongemma-multiagent-router"
 
 tokenizer = AutoTokenizer.from_pretrained(model_name)
+
 model = AutoModelForCausalLM.from_pretrained(
-    model_name,
-    device_map="auto",
-    torch_dtype="auto"
+    model_name
 )
 
 # -----------------------
 # Agents
 # -----------------------
 def technical_support_agent(issue_type: str, priority: str) -> str:
-    return f"🛠️ Routing to Technical Support: {issue_type} ({priority})"
+    return f"🛠️ Technical Support → {issue_type} ({priority})"
 
 def billing_agent(request_type: str, urgency: str) -> str:
-    return f"💰 Routing to Billing: {request_type} ({urgency})"
+    return f"💰 Billing → {request_type} ({urgency})"
 
 def product_info_agent(query_type: str, category: str) -> str:
-    return f"📦 Routing to Product Info: {query_type} ({category})"
+    return f"📦 Product Info → {query_type} ({category})"
 
-# Tool schemas
 AGENT_TOOLS = [
     get_json_schema(technical_support_agent),
     get_json_schema(billing_agent),
@@ -37,7 +34,7 @@ AGENT_TOOLS = [
 SYSTEM_MSG = "You are an intelligent routing agent that directs customer queries to the appropriate specialized agent."
 
 # -----------------------
-# Core inference
+# Inference
 # -----------------------
 def route_query(user_query: str):
 
@@ -50,28 +47,24 @@ def route_query(user_query: str):
         messages,
         tools=AGENT_TOOLS,
         add_generation_prompt=True,
-        return_dict=True,
         return_tensors="pt"
     )
 
-    inputs = {k: v.to(model.device) for k, v in inputs.items()}
-
     outputs = model.generate(
-        **inputs,
-        max_new_tokens=128,
-        pad_token_id=tokenizer.eos_token_id
+        inputs,
+        max_new_tokens=100
     )
 
     result = tokenizer.decode(
-        outputs[0][len(inputs["input_ids"][0]):],
+        outputs[0],
         skip_special_tokens=True
     )
 
-    return result
+    return result.split("assistant")[-1].strip()
 
 
 # -----------------------
-# Chatbot logic
+# Chat UI
 # -----------------------
 def chat_fn(message, history):
     response = route_query(message)
@@ -79,19 +72,21 @@ def chat_fn(message, history):
     return history, history
 
 
-# -----------------------
-# UI
-# -----------------------
 with gr.Blocks() as demo:
-    gr.Markdown("## 🤖 Multi-Agent Router Chatbot")
-    gr.Markdown("Ask anything about billing, product, or technical issues.")
-
-    chatbot = gr.Chatbot()
-    msg = gr.Textbox(placeholder="Type your query here...")
-    clear = gr.Button("Clear")
+    gr.Markdown("## 🤖 Multi-Agent Router (Fast CPU Demo)")
+    
+    chatbot = gr.Chatbot(height=400)
+    msg = gr.Textbox(placeholder="Ask something...")
+    
+    gr.Examples(
+        examples=[
+            "My app crashes when uploading files",
+            "I want a refund",
+            "What features are in premium plan?"
+        ],
+        inputs=msg
+    )
 
     msg.submit(chat_fn, [msg, chatbot], [chatbot, chatbot])
-    clear.click(lambda: None, None, chatbot, queue=False)
 
-# Launch
 demo.launch()