llm_server/routes/ooba_request_handler.py

@@ -16,7 +16,7 @@ class OobaRequestHandler(RequestHandler):
     def handle_request(self, return_ok: bool = True):
         assert not self.used
         if self.offline:
-            msg = f'{self.selected_model} is not a valid model choice.'
+            msg = 'The model you requested is not a valid choice. Please retry your query.'
             print(msg)
             self.handle_error(msg)
 

server.py

@@ -1,5 +1,3 @@
-from llm_server.routes.queue import priority_queue
-
 try:
     import gevent.monkey
 
@@ -26,6 +24,7 @@ from llm_server.routes.server_error import handle_server_error
 from llm_server.routes.v1 import bp
 from llm_server.sock import init_socketio
 
+# TODO: return an `error: True`, error code, and error message rather than just a formatted message
 # TODO: what happens when all backends are offline? What about the "online" key in the stats page?
 # TODO: redis SCAN vs KEYS??
 # TODO: implement blind RRD controlled via header and only used when there is a queue on the primary backend(s)