more

app.py

@@ -46,7 +46,7 @@ def respond(message, max_new_tokens, temperature, system_message=""):
 
         # Stream response
         text_streamer = TextStreamer(tokenizer, skip_prompt=True)
-        response = model.generate(
+        model.generate(
             input_ids=input_ids,
             max_new_tokens=max_new_tokens,
             temperature=temperature,
@@ -54,7 +54,14 @@ def respond(message, max_new_tokens, temperature, system_message=""):
             streamer=text_streamer,
         )
 
-        return response
+        # Get the response generated by the model
+        # Retrieve text from the output stream (assuming this works with your setup)
+        generated_text = text_streamer.generated_text  # This assumes the `TextStreamer` accumulates the generated text
+
+        # Debug: Show the response text
+        print("[DEBUG] Generated Text:", generated_text)
+
+        return generated_text
     except Exception as e:
         # Debug: Log errors
         print("[ERROR]", str(e))