README.md

@@ -8,5 +8,6 @@ sdk_version: 3.35.2
 app_file: app.py
 pinned: false
 ---
+NB: Need a CPU Upgrade (32GB RAM) instance to run on a huggingface space
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -26,10 +26,12 @@ def predict0(prompt, bot):
         user_prompt = prompt
         generator = generate(llm, generation_config, system_prompt, user_prompt.strip())
         print(assistant_prefix, end=" ", flush=True)
+
+        response = ""
         for word in generator:
             print(word, end="", flush=True)
         print("")
-        response = word
+        response += word
         logger.debug(f"{response=}")
     except Exception as exc:
         logger.error(exc)