Fixes descriptions in the Inference APIs (#5566)

kosabogi · leemthompo · davidkyle · web-flow · commit 518ccd9a1ab3 · 2025-11-04T15:42:28.000+01:00
* Improves descriptions in the Inference APIs

* Update specification/inference/delete/DeleteRequest.ts

Co-authored-by: Liam Thompson &lt;leemthompo@gmail.com&gt;

* Update specification/inference/put/PutRequest.ts

Co-authored-by: David Kyle &lt;david.kyle@elastic.co&gt;

* Fixes chunking_settings object descriptions

* Fixes oxford comma

* Formatting fix

---------

Co-authored-by: Liam Thompson &lt;leemthompo@gmail.com&gt;
Co-authored-by: David Kyle &lt;david.kyle@elastic.co&gt;
diff --git a/specification/inference/_types/CommonTypes.ts b/specification/inference/_types/CommonTypes.ts
@@ -391,7 +391,7 @@ export class AlibabaCloudTaskSettings {
 export enum AlibabaCloudTaskType {
   completion,
   rerank,
-  space_embedding,
+  sparse_embedding,
   text_embedding
 }
 
diff --git a/specification/inference/delete/DeleteRequest.ts b/specification/inference/delete/DeleteRequest.ts
@@ -23,6 +23,7 @@ import { TaskType } from '@inference/_types/TaskType'
 
 /**
  * Delete an inference endpoint
+ * This API requires the manage_inference cluster privilege (the built-in `inference_admin` role grants this privilege).
  * @rest_spec_name inference.delete
  * @availability stack since=8.11.0 stability=stable visibility=public
  * @availability serverless stability=stable visibility=public
diff --git a/specification/inference/get/GetRequest.ts b/specification/inference/get/GetRequest.ts
@@ -23,6 +23,7 @@ import { TaskType } from '@inference/_types/TaskType'
 
 /**
  * Get an inference endpoint
+ * This API requires the `monitor_inference` cluster privilege (the built-in `inference_admin` and `inference_user` roles grant this privilege).
  * @rest_spec_name inference.get
  * @availability stack since=8.11.0 stability=stable visibility=public
  * @availability serverless stability=stable visibility=public
diff --git a/specification/inference/put_alibabacloud/PutAlibabaCloudRequest.ts b/specification/inference/put_alibabacloud/PutAlibabaCloudRequest.ts
@@ -65,6 +65,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `sparse_embedding` or `text_embedding` task types.
+     * Not applicable to the `rerank` or `completion` task types.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_amazonbedrock/PutAmazonBedrockRequest.ts b/specification/inference/put_amazonbedrock/PutAmazonBedrockRequest.ts
@@ -68,6 +68,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `completion` task type.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_amazonsagemaker/PutAmazonSageMakerRequest.ts b/specification/inference/put_amazonsagemaker/PutAmazonSageMakerRequest.ts
@@ -65,6 +65,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `sparse_embedding` or `text_embedding` task types.
+     * Not applicable to the `rerank`, `completion`, or `chat_completion` task types.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_anthropic/PutAnthropicRequest.ts b/specification/inference/put_anthropic/PutAnthropicRequest.ts
@@ -26,7 +26,6 @@ import {
   AnthropicTaskSettings,
   AnthropicTaskType
 } from '@inference/_types/CommonTypes'
-import { InferenceChunkingSettings } from '@inference/_types/Services'
 
 /**
  * Create an Anthropic inference endpoint.
@@ -64,11 +63,6 @@ export interface Request extends RequestBase {
     timeout?: Duration
   }
   body: {
-    /**
-     * The chunking configuration object.
-     * @ext_doc_id inference-chunking
-     */
-    chunking_settings?: InferenceChunkingSettings
     /**
      * The type of service supported for the specified task type. In this case, `anthropic`.
      */
diff --git a/specification/inference/put_azureaistudio/PutAzureAiStudioRequest.ts b/specification/inference/put_azureaistudio/PutAzureAiStudioRequest.ts
@@ -65,6 +65,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `rerank` or `completion` task types.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_azureopenai/PutAzureOpenAiRequest.ts b/specification/inference/put_azureopenai/PutAzureOpenAiRequest.ts
@@ -73,6 +73,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `completion` task type.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_cohere/PutCohereRequest.ts b/specification/inference/put_cohere/PutCohereRequest.ts
@@ -65,6 +65,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `rerank` or `completion` task type.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_contextualai/PutContextualAiRequest.ts b/specification/inference/put_contextualai/PutContextualAiRequest.ts
@@ -25,7 +25,6 @@ import {
   ContextualAIServiceType,
   ContextualAITaskSettings
 } from '@inference/_types/CommonTypes'
-import { InferenceChunkingSettings } from '@inference/_types/Services'
 import { TaskTypeContextualAI } from '@inference/_types/TaskType'
 
 /**
@@ -65,11 +64,6 @@ export interface Request extends RequestBase {
     timeout?: Duration
   }
   body: {
-    /**
-     * The chunking configuration object.
-     * @ext_doc_id inference-chunking
-     */
-    chunking_settings?: InferenceChunkingSettings
     /**
      * The type of service supported for the specified task type. In this case, `contextualai`.
      */
diff --git a/specification/inference/put_custom/PutCustomRequest.ts b/specification/inference/put_custom/PutCustomRequest.ts
@@ -96,6 +96,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `sparse_embedding` or `text_embedding` task types.
+     * Not applicable to the `rerank` or `completion` task types.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_deepseek/PutDeepSeekRequest.ts b/specification/inference/put_deepseek/PutDeepSeekRequest.ts
@@ -24,7 +24,6 @@ import {
   DeepSeekServiceSettings,
   DeepSeekServiceType
 } from '@inference/_types/CommonTypes'
-import { InferenceChunkingSettings } from '@inference/_types/Services'
 import { TaskTypeDeepSeek } from '@inference/_types/TaskType'
 
 /**
@@ -62,11 +61,6 @@ export interface Request extends RequestBase {
     timeout?: Duration
   }
   body: {
-    /**
-     * The chunking configuration object.
-     * @ext_doc_id inference-chunking
-     */
-    chunking_settings?: InferenceChunkingSettings
     /**
      * The type of service supported for the specified task type. In this case, `deepseek`.
      */
diff --git a/specification/inference/put_elasticsearch/PutElasticsearchRequest.ts b/specification/inference/put_elasticsearch/PutElasticsearchRequest.ts
@@ -80,7 +80,7 @@ export interface Request extends RequestBase {
     /**
      * The chunking configuration object.
      * Applies only to the `sparse_embedding` and `text_embedding` task types.
-     * Not applicable to the `rerank`, `completion`, or `chat_completion` task types.
+     * Not applicable to the `rerank` task type.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_googleaistudio/PutGoogleAiStudioRequest.ts b/specification/inference/put_googleaistudio/PutGoogleAiStudioRequest.ts
@@ -64,6 +64,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `completion` task type.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_googlevertexai/PutGoogleVertexAiRequest.ts b/specification/inference/put_googlevertexai/PutGoogleVertexAiRequest.ts
@@ -65,6 +65,8 @@ export interface Request extends RequestBase {
   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
diff --git a/specification/inference/put_hugging_face/PutHuggingFaceRequest.ts b/specification/inference/put_hugging_face/PutHuggingFaceRequest.ts
@@ -101,6 +101,8 @@ export interface Request extends RequestBase {
   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
diff --git a/specification/inference/put_jinaai/PutJinaAiRequest.ts b/specification/inference/put_jinaai/PutJinaAiRequest.ts
@@ -68,6 +68,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `rerank` task type.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_llama/PutLlamaRequest.ts b/specification/inference/put_llama/PutLlamaRequest.ts
@@ -64,6 +64,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `completion` or `chat_completion` task types.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_mistral/PutMistralRequest.ts b/specification/inference/put_mistral/PutMistralRequest.ts
@@ -64,6 +64,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `completion` or `chat_completion` task types.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_openai/PutOpenAiRequest.ts b/specification/inference/put_openai/PutOpenAiRequest.ts
@@ -66,6 +66,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `completion` or `chat_completion` task types.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git a/specification/inference/put_voyageai/PutVoyageAIRequest.ts b/specification/inference/put_voyageai/PutVoyageAIRequest.ts
@@ -67,6 +67,8 @@ export interface Request extends RequestBase {
   body: {
     /**
      * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `rerank` task type.
      * @ext_doc_id inference-chunking
      */
     chunking_settings?: InferenceChunkingSettings
diff --git 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.
@@ -63,6 +64,13 @@ export interface Request extends RequestBase {
     timeout?: Duration
   }
   body: {
+    /**
+     * The chunking configuration object.
+     * Applies only to the `text_embedding` task type.
+     * Not applicable to the `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`.
      */

Original file line number	Diff line number	Diff line change
`@@ -391,7 +391,7 @@ export class AlibabaCloudTaskSettings {`
`391`	`391`	`export enum AlibabaCloudTaskType {`
`392`	`392`	`completion,`
`393`	`393`	`rerank,`
`394`		`- space_embedding,`
	`394`	`+ sparse_embedding,`
`395`	`395`	`text_embedding`
`396`	`396`	`}`
`397`	`397`