diff --git a/deploy/macae-continer-oc.json b/deploy/macae-continer-oc.json
index 40c676ebe..987f67615 100644
--- a/deploy/macae-continer-oc.json
+++ b/deploy/macae-continer-oc.json
@@ -5,8 +5,8 @@
   "metadata": {
     "_generator": {
       "name": "bicep",
-      "version": "0.33.93.31351",
-      "templateHash": "9524414973084491660"
+      "version": "0.34.44.8038",
+      "templateHash": "4215337980430060491"
     }
   },
   "parameters": {
@@ -63,7 +63,7 @@
         }
       },
       "defaultValue": {
-        "gpt4oCapacity": 50,
+        "gpt4oCapacity": 5,
         "containerAppSize": {
           "cpu": "2.0",
           "memory": "4.0Gi",
@@ -376,12 +376,12 @@
       },
       "dependsOn": [
         "appInsights",
-        "cosmos::autogenDb",
         "containerAppEnv",
         "cosmos",
-        "openai::gpt4o",
+        "cosmos::autogenDb",
         "cosmos::autogenDb::memoryContainer",
         "openai",
+        "openai::gpt4o",
         "pullIdentity"
       ],
       "metadata": {
diff --git a/deploy/macae-continer.bicep b/deploy/macae-continer.bicep
index 407879b75..0678a28c1 100644
--- a/deploy/macae-continer.bicep
+++ b/deploy/macae-continer.bicep
@@ -22,7 +22,7 @@ param resourceSize {
     maxReplicas: int
   }
 } = {
-  gpt4oCapacity: 50
+  gpt4oCapacity: 5
   containerAppSize: {
     cpu: '2.0'
     memory: '4.0Gi'
diff --git a/deploy/macae-dev.bicep b/deploy/macae-dev.bicep
index 5157fa92f..111bf229a 100644
--- a/deploy/macae-dev.bicep
+++ b/deploy/macae-dev.bicep
@@ -36,7 +36,7 @@ resource openai 'Microsoft.CognitiveServices/accounts@2023-10-01-preview' = {
     name: 'gpt-4o'
     sku: {
       name: 'GlobalStandard'
-      capacity: 15
+      capacity: 5
     }
     properties: {
       model: {
diff --git a/deploy/macae-large.bicepparam b/deploy/macae-large.bicepparam
index 3e88f4452..27665f125 100644
--- a/deploy/macae-large.bicepparam
+++ b/deploy/macae-large.bicepparam
@@ -1,7 +1,7 @@
 using './macae.bicep'
 
 param resourceSize = {
-  gpt4oCapacity: 50
+  gpt4oCapacity: 5
   containerAppSize: {
     cpu: '2.0'
     memory: '4.0Gi'
diff --git a/deploy/macae-mini.bicepparam b/deploy/macae-mini.bicepparam
index ee3d65127..6aa9dd06d 100644
--- a/deploy/macae-mini.bicepparam
+++ b/deploy/macae-mini.bicepparam
@@ -1,7 +1,7 @@
 using './macae.bicep'
 
 param resourceSize = {
-  gpt4oCapacity: 15
+  gpt4oCapacity: 5
   containerAppSize: {
     cpu: '1.0'
     memory: '2.0Gi'
diff --git a/deploy/macae.bicep b/deploy/macae.bicep
index bfa56c9a1..a6bf66f4b 100644
--- a/deploy/macae.bicep
+++ b/deploy/macae.bicep
@@ -25,7 +25,7 @@ param resourceSize {
     maxReplicas: int
   }
 } = {
-  gpt4oCapacity: 50
+  gpt4oCapacity: 5
   containerAppSize: {
     cpu: '2.0'
     memory: '4.0Gi'
diff --git a/infra/macae-dev.bicep b/infra/macae-dev.bicep
index 5157fa92f..111bf229a 100644
--- a/infra/macae-dev.bicep
+++ b/infra/macae-dev.bicep
@@ -36,7 +36,7 @@ resource openai 'Microsoft.CognitiveServices/accounts@2023-10-01-preview' = {
     name: 'gpt-4o'
     sku: {
       name: 'GlobalStandard'
-      capacity: 15
+      capacity: 5
     }
     properties: {
       model: {
diff --git a/infra/macae-large.bicepparam b/infra/macae-large.bicepparam
index 3e88f4452..27665f125 100644
--- a/infra/macae-large.bicepparam
+++ b/infra/macae-large.bicepparam
@@ -1,7 +1,7 @@
 using './macae.bicep'
 
 param resourceSize = {
-  gpt4oCapacity: 50
+  gpt4oCapacity: 5
   containerAppSize: {
     cpu: '2.0'
     memory: '4.0Gi'
diff --git a/infra/macae-mini.bicepparam b/infra/macae-mini.bicepparam
index ee3d65127..6aa9dd06d 100644
--- a/infra/macae-mini.bicepparam
+++ b/infra/macae-mini.bicepparam
@@ -1,7 +1,7 @@
 using './macae.bicep'
 
 param resourceSize = {
-  gpt4oCapacity: 15
+  gpt4oCapacity: 5
   containerAppSize: {
     cpu: '1.0'
     memory: '2.0Gi'
diff --git a/infra/macae.bicep b/infra/macae.bicep
index bfa56c9a1..a6bf66f4b 100644
--- a/infra/macae.bicep
+++ b/infra/macae.bicep
@@ -25,7 +25,7 @@ param resourceSize {
     maxReplicas: int
   }
 } = {
-  gpt4oCapacity: 50
+  gpt4oCapacity: 5
   containerAppSize: {
     cpu: '2.0'
     memory: '4.0Gi'
diff --git a/infra/main.bicep b/infra/main.bicep
index fecb5c751..eb757f864 100644
--- a/infra/main.bicep
+++ b/infra/main.bicep
@@ -22,7 +22,7 @@ param resourceSize {
     maxReplicas: int
   }
 } = {
-  gpt4oCapacity: 1
+  gpt4oCapacity: 5
   containerAppSize: {
     cpu: '2.0'
     memory: '4.0Gi'
diff --git a/infra/main.json b/infra/main.json
index 9f6864aae..555d0f4c4 100644
--- a/infra/main.json
+++ b/infra/main.json
@@ -6,7 +6,7 @@
     "_generator": {
       "name": "bicep",
       "version": "0.34.44.8038",
-      "templateHash": "2906892014954666053"
+      "templateHash": "3131309205289650458"
     }
   },
   "parameters": {
@@ -63,7 +63,7 @@
         }
       },
       "defaultValue": {
-        "gpt4oCapacity": 1,
+        "gpt4oCapacity": 5,
         "containerAppSize": {
           "cpu": "2.0",
           "memory": "4.0Gi",