diff --git a/release/deployment/docker-compose/conf/model_config.yaml b/release/deployment/docker-compose/conf/model_config.yaml
index 2aab6ca79..018424dec 100644
--- a/release/deployment/docker-compose/conf/model_config.yaml
+++ b/release/deployment/docker-compose/conf/model_config.yaml
@@ -29,3 +29,15 @@ models:
           min: "0.001"
           max: "1.0"
           default_val: "0.7"
+        - name: "thinking"
+          label: "thinking"
+          desc: "Enable thinking mode for reasoning models that support extended thinking capabilities."
+          type: "boolean"
+          default_val: "false"
+        - name: "thinking_budget_tokens"
+          label: "thinking_budget_tokens"
+          desc: "Maximum number of tokens for thinking content output when thinking mode is enabled."
+          type: "int"
+          min: "1024"
+          max: "65536"
+          default_val: "4096"
diff --git a/release/deployment/helm-chart/umbrella/conf/model_config.yaml b/release/deployment/helm-chart/umbrella/conf/model_config.yaml
index 2aab6ca79..018424dec 100644
--- a/release/deployment/helm-chart/umbrella/conf/model_config.yaml
+++ b/release/deployment/helm-chart/umbrella/conf/model_config.yaml
@@ -29,3 +29,15 @@ models:
           min: "0.001"
           max: "1.0"
           default_val: "0.7"
+        - name: "thinking"
+          label: "thinking"
+          desc: "Enable thinking mode for reasoning models that support extended thinking capabilities."
+          type: "boolean"
+          default_val: "false"
+        - name: "thinking_budget_tokens"
+          label: "thinking_budget_tokens"
+          desc: "Maximum number of tokens for thinking content output when thinking mode is enabled."
+          type: "int"
+          min: "1024"
+          max: "65536"
+          default_val: "4096"