diff --git a/serverless/pages/ml-nlp-auto-scale.mdx b/serverless/pages/ml-nlp-auto-scale.mdx
index 67c01ff..7723440 100644
--- a/serverless/pages/ml-nlp-auto-scale.mdx
+++ b/serverless/pages/ml-nlp-auto-scale.mdx
@@ -125,7 +125,7 @@ The maximum number of threads that can be claimed depends on the hardware your a
 
 #### Adaptive resources disabled
 
-| Level  | Allocations                                             | Threads                | vCPUs                                                |
+| Level  | Allocations                                             | Threads                | vCUs                                                |
 |--------|---------------------------------------------------------|------------------------|------------------------------------------------------|
 | Low    | 1 statically                                             | Always 2               | 2                                                     |
 | Medium | 2 statically (if threads=16)                             | Maximum (for example, 16)         | 32                                                    |