diff --git a/specification/inference/_types/CommonTypes.ts b/specification/inference/_types/CommonTypes.ts
index 948548d0cc..d4a1004437 100644
--- a/specification/inference/_types/CommonTypes.ts
+++ b/specification/inference/_types/CommonTypes.ts
@@ -1695,6 +1695,7 @@ export class WatsonxServiceSettings {
 
 export enum WatsonxTaskType {
   text_embedding,
+  rerank,
   chat_completion,
   completion
 }
diff --git a/specification/inference/put/PutRequest.ts b/specification/inference/put/PutRequest.ts
index 2949bbd4d6..3201552235 100644
--- a/specification/inference/put/PutRequest.ts
+++ b/specification/inference/put/PutRequest.ts
@@ -49,7 +49,7 @@ import { TaskType } from '@inference/_types/TaskType'
  * * Mistral (`chat_completion`, `completion`, `text_embedding`)
  * * OpenAI (`chat_completion`, `completion`, `text_embedding`)
  * * VoyageAI (`rerank`, `text_embedding`)
- * * Watsonx inference integration (`text_embedding`)
+ * * Watsonx (`chat_completion`, `completion`, `rerank`, `text_embedding`)
  * @rest_spec_name inference.put
  * @availability stack since=8.11.0 stability=stable visibility=public
  * @availability serverless stability=stable visibility=public
diff --git a/specification/inference/put_watsonx/PutWatsonxRequest.ts b/specification/inference/put_watsonx/PutWatsonxRequest.ts
index 6d28e7e7f2..2d13d0ad6e 100644
--- a/specification/inference/put_watsonx/PutWatsonxRequest.ts
+++ b/specification/inference/put_watsonx/PutWatsonxRequest.ts
@@ -25,6 +25,7 @@ import {
   WatsonxServiceType,
   WatsonxTaskType
 } from '@inference/_types/CommonTypes'
+import { InferenceChunkingSettings } from '@inference/_types/Services'
 
 /**
  * Create a Watsonx inference endpoint.
@@ -65,6 +66,13 @@ export interface Request extends RequestBase {
     timeout?: Duration
   }
   body: {
+    /**
+     * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `rerank`, `completion` or `chat_completion` task types.
+     * @ext_doc_id inference-chunking
+     */
+    chunking_settings?: InferenceChunkingSettings
     /**
      * The type of service supported for the specified task type. In this case, `watsonxai`.
      */