diff --git a/async-openai/src/types/chat.rs b/async-openai/src/types/chat.rs
index d9373db6..46c0d7dc 100644
--- a/async-openai/src/types/chat.rs
+++ b/async-openai/src/types/chat.rs
@@ -618,6 +618,8 @@ pub enum ServiceTier {
     Flex,
     Scale,
     Priority,
+    #[serde(rename = "on_demand")]
+    OnDemand,
 }
 
 #[derive(Clone, Serialize, Debug, Deserialize, PartialEq)]
@@ -627,6 +629,8 @@ pub enum ServiceTierResponse {
     Default,
     Flex,
     Priority,
+    #[serde(rename = "on_demand")]
+    OnDemand,
 }
 
 #[derive(Clone, Serialize, Debug, Deserialize, PartialEq)]
diff --git a/async-openai/src/types/responses.rs b/async-openai/src/types/responses.rs
index abccf301..b8e3a2a3 100644
--- a/async-openai/src/types/responses.rs
+++ b/async-openai/src/types/responses.rs
@@ -334,6 +334,9 @@ pub enum ServiceTier {
     Flex,
     Scale,
     Priority,
+
+    #[serde(rename = "on_demand")]
+    OnDemand,
 }
 
 /// Truncation strategies.