Update deploy.json

deploy.json

@@ -1,13 +1,11 @@
 {
   "parameters": {
-    "max_total_tokens": 2048,
-    "max_input_length": 1024,
-    "max_batch_total_tokens": 8192,
-    "max_concurrent_requests": 1,
-    "max_best_of": 1,
-    "max_stop_sequences": 4,
-    "max_batch_size": 1,
-    "waiting_served_ratio": 1.2
+    "max_total_tokens": 4096,          // Increase from 2048
+    "max_input_length": 2048,          // Increase from 1024
+    "max_batch_total_tokens": 16384,   // Increase from 8192
+    "max_concurrent_requests": 2,       // Increase from 1
+    "max_batch_size": 2,               // Increase from 1
+    "waiting_served_ratio": 0.8        // Decrease from 1.2
   },
   "hardware": {
     "task_type": "text-generation",
@@ -17,7 +15,7 @@
     "distributed_setup": false
   },
   "framework_type": "pytorch",
-  "torch_compile": false,
+  "torch_compile": true,
   "trust_remote_code": true,
-  "disable_custom_kernels": true
+  "disable_custom_kernels": false
 }