cache

app.py

@@ -115,6 +115,7 @@ def run(hf_token, service, game, functionality, nlp_command):
                 timecost = calling_end.timestamp() - calling_start.timestamp()
                 print(f"calling ai ends at {calling_end}, costs {timecost:.2f}s")
                 outp = json.loads(res)
+                outp["cache"] = False
                 cache.add_cache(nlp_command, outp)
             # add proxy version info to the output
             outp["proxy-version"] = proxy_version