fix(halo): only one model

systems/x86_64-linux/halo/llama-server.nix

@@ -29,7 +29,7 @@
         "--host 0.0.0.0"
         "--port 8000"
         "--models-preset ${./models.ini}"
-        "--models-max 2"
+        "--models-max 1"
       ];
       Restart = "on-failure";
       RestartSec = 10;

systems/x86_64-linux/halo/models.ini

@@ -11,7 +11,6 @@ cache-type-k        = bf16
 cache-type-v        = bf16
 mmap                = false
 no-context-shift    = true
-chat-template-kwargs = {"preserve_thinking": true}
 fit                 = on
 c                   = 131072
 
@@ -27,3 +26,4 @@ min-p               = 0.0
 parallel            = 2
 c                   = 524288
 load-on-startup     = true
+chat-template-kwargs = {"preserve_thinking": true}