Update app.py

app.py

@@ -75,20 +75,23 @@ def bot_streaming(message, history, temperature, top_k, max_new_tokens):
         "top_k": top_k,
     }
     
-    # Generate from the VLM  
-    thread = Thread(target=vlm.generate, kwargs=generation_kwargs)
-    thread.start()
+    # Generate from the VLM
+    buffer = vlm.generate(
+    **generation_kwargs
+    )
+    # thread = Thread(target=vlm.generate, kwargs=generation_kwargs)
+    # thread.start()
 
-    buffer = ""
-    output_started = False
-    for new_text in streamer:
-        if not output_started:
-            if "<|assistant|>\n" in new_text:
-                output_started = True
-            continue
-        buffer += new_text
-        if len(buffer) > 1:
-            yield buffer
+    # buffer = ""
+    # output_started = False
+    # for new_text in streamer:
+    #     if not output_started:
+    #         if "<|assistant|>\n" in new_text:
+    #             output_started = True
+    #         continue
+    #     buffer += new_text
+    #     if len(buffer) > 1:
+    #         yield buffer
 
     prompt_builder.add_turn(role="gpt", message=buffer)