Update app.py

app.py

@@ -51,7 +51,7 @@ llm = None
 llm_model = None
 
 @spaces.GPU(duration=120)
-async def respond(
+def respond(
     message,
     history: list[tuple[str, str]],
     model,
@@ -61,9 +61,9 @@ async def respond(
     top_p,
     top_k,
     repeat_penalty,
-    Template,
+    template,
 ):
-    chat_template = MessagesFormatterType.GEMMA_2
+    chat_template = MessagesFormatterType[template]
 
     global llm
     global llm_model
@@ -118,12 +118,9 @@ async def respond(
     )
     
     outputs = ""
-    try:
-        async for output in stream:
-            outputs += output
-            yield outputs
-    except StopAsyncIteration:
-        pass  # Handle end of iteration gracefully
+    for output in stream:
+        outputs += output
+        yield outputs
 
 description = """<p align="center">Defaults to Oumuamua-7b-RP (you can switch to other models from additional inputs)</p>
 <p><center>