small changes

app.py

@@ -34,7 +34,8 @@ def respond(message, history, system_message, max_tokens, temperature, top_p):
         tokenize=True,
         add_generation_prompt=True,  # Required for generation
         return_tensors="pt",
-    ).to("cuda" if torch.cuda.is_available() else "cpu")
+    )
+    input_ids = inputs.to("cuda" if torch.cuda.is_available() else "cpu")
 
     # Initialize a TextStreamer for streaming output
     text_streamer = TextStreamer(tokenizer, skip_prompt=True)
@@ -42,7 +43,7 @@ def respond(message, history, system_message, max_tokens, temperature, top_p):
     # Generate the model's response
     response = ""
     for token_ids in model.generate(
-        input_ids=inputs.input_ids,
+        input_ids=input_ids,  # Use the tensor directly
         streamer=text_streamer,
         max_new_tokens=max_tokens,
         use_cache=True,
@@ -66,4 +67,4 @@ demo = gr.ChatInterface(
 )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(share=True)