[DOCS] Adjusts note on minimum node size for ELSER and third-party models (#2447)

Co-authored-by: David Roberts <[email protected]>

Author: szabosteve

docs/en/stack/ml/nlp/ml-nlp-elser.asciidoc

@@ -36,9 +36,14 @@ weights at index time, and to search against later.
 To use ELSER, you must have the {subscriptions}[appropriate subscription] level 
 for semantic search or the trial period activated.
 
-NOTE: The minimum dedicated ML node size for deploying and using 
-the ELSER model is 4 GB. This is a minimum and better performance
-can be achieved by using bigger ML nodes.
+NOTE: The minimum dedicated ML node size for deploying and using the ELSER model 
+is 4 GB in Elasticsearch Service if 
+{cloud}/ec-autoscaling.html[deployment autoscaling] is turned off. Turning on 
+autoscaling is recommended because it allows your deployment to dynamically 
+adjust resources based on demand. Better performance can be achieved by using 
+more allocations or more threads per allocation, which requires bigger ML nodes. 
+Autoscaling provides bigger nodes when required. If autoscaling is turned off, 
+you must provide suitably sized nodes yourself.
 
 
 [discrete]

docs/en/stack/ml/nlp/ml-nlp-model-ref.asciidoc

@@ -9,6 +9,15 @@
 :frontmatter-tags-content-type: [reference] 
 :frontmatter-tags-user-goals: [analyze]
 
+The minimum dedicated ML node size for deploying and using the ELSER model 
+is 16 GB in Elasticsearch Service if 
+{cloud}/ec-autoscaling.html[deployment autoscaling] is turned off. Turning on 
+autoscaling is recommended because it allows your deployment to dynamically 
+adjust resources based on demand. Better performance can be achieved by using 
+more allocations or more threads per allocation, which requires bigger ML nodes. 
+Autoscaling provides bigger nodes when required. If autoscaling is turned off, 
+you must provide suitably sized nodes yourself.
+
 The {stack-ml-features} support transformer models that conform to the standard
 BERT model interface and use the WordPiece tokenization algorithm.