Update app.py

app.py

@@ -75,12 +75,14 @@ def generate(prompt, history=[], temperature=0.1, max_new_tokens=10000, top_p=0.
         return
 
     formatted_prompt = format_prompt(prompt, history)
+    output_accumulated = ""
     try:
         stream = client.text_generation(formatted_prompt, temperature=temperature, max_new_tokens=min(max_new_tokens, available_tokens),
                                         top_p=top_p, repetition_penalty=repetition_penalty, do_sample=True, seed=42, stream=True)
         for response in stream:
-            output = response['generated_text'] if 'generated_text' in response else str(response)
-            yield f"{output}\n\n---\nTotal tokens used: {total_tokens_used}"
+            output_part = response['generated_text'] if 'generated_text' in response else str(response)
+            output_accumulated += output_part
+            yield output_accumulated + f"\n\n---\nTotal tokens used: {total_tokens_used}"
     except Exception as e:
         yield f"Error: {str(e)}\nTotal tokens used: {total_tokens_used}"