Update app.py

app.py

@@ -7,7 +7,7 @@ from huggingface_hub import InferenceClient
 import os
 
 # Initialize Qwen client
-qwen_client = InferenceClient("deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B")
+qwen_client = InferenceClient(os.environ.get("HF_TOKEN"))
 
 # Model and ONNX setup
 HG_MODEL = "livekit/turn-detector"
@@ -84,6 +84,7 @@ def respond(
     # Generate raw response without any processing
     full_response = ""
     stream = qwen_client.chat.completions.create(
+        model="deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B", 
         messages=messages,
         max_tokens=max_tokens,
         stream=True,
@@ -93,11 +94,10 @@ def respond(
     )
 
     for chunk in stream:
-        print(chunk.choices[0].delta.content)
+        full_response += chunk.choices[0].delta.content
         yield chunk.choices[0].delta.content  # Send raw unmodified response to Gradio
 
     # This will match both console and Gradio output
-    print(f"Final response: {full_response}")
 # Create Gradio interface
 demo = gr.ChatInterface(
     respond,