elasticsearch 9.0.4__py3-none-any.whl → 9.1.1__py3-none-any.whl

elasticsearch/_sync/client/inference.py

@@ -47,7 +47,7 @@ class InferenceClient(NamespacedClient):
           <p>Perform completion inference on the service</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-inference>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-inference>`_
 
         :param inference_id: The inference Id
         :param input: Inference input. Either a string or an array of strings.
@@ -123,7 +123,7 @@ class InferenceClient(NamespacedClient):
           <p>Delete an inference endpoint</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-delete>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-delete>`_
 
         :param inference_id: The inference identifier.
         :param task_type: The task type
@@ -197,7 +197,7 @@ class InferenceClient(NamespacedClient):
           <p>Get an inference endpoint</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-get>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-get>`_
 
         :param task_type: The task type
         :param inference_id: The inference Id
@@ -235,7 +235,7 @@ class InferenceClient(NamespacedClient):
         )
 
     @_rewrite_parameters(
-        body_fields=("input", "query", "task_settings"),
+        body_fields=("input", "input_type", "query", "task_settings"),
     )
     def inference(
         self,
@@ -257,6 +257,7 @@ class InferenceClient(NamespacedClient):
         error_trace: t.Optional[bool] = None,
         filter_path: t.Optional[t.Union[str, t.Sequence[str]]] = None,
         human: t.Optional[bool] = None,
+        input_type: t.Optional[str] = None,
         pretty: t.Optional[bool] = None,
         query: t.Optional[str] = None,
         task_settings: t.Optional[t.Any] = None,
@@ -277,13 +278,22 @@ class InferenceClient(NamespacedClient):
           </blockquote>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-inference>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-inference>`_
 
         :param inference_id: The unique identifier for the inference endpoint.
         :param input: The text on which you want to perform the inference task. It can
             be a single string or an array. > info > Inference endpoints for the `completion`
             task type currently only support a single string as input.
         :param task_type: The type of inference task that the model performs.
+        :param input_type: Specifies the input data type for the text embedding model.
+            The `input_type` parameter only applies to Inference Endpoints with the `text_embedding`
+            task type. Possible values include: * `SEARCH` * `INGEST` * `CLASSIFICATION`
+            * `CLUSTERING` Not all services support all values. Unsupported values will
+            trigger a validation exception. Accepted values depend on the configured
+            inference service, refer to the relevant service-specific documentation for
+            more info. > info > The `input_type` parameter specified on the root level
+            of the request body will take precedence over the `input_type` parameter
+            specified in `task_settings`.
         :param query: The query input, which is required only for the `rerank` task.
             It is not required for other tasks.
         :param task_settings: Task settings for the individual inference request. These
@@ -322,6 +332,8 @@ class InferenceClient(NamespacedClient):
         if not __body:
             if input is not None:
                 __body["input"] = input
+            if input_type is not None:
+                __body["input_type"] = input_type
             if query is not None:
                 __body["query"] = query
             if task_settings is not None:
@@ -379,24 +391,27 @@ class InferenceClient(NamespacedClient):
           <ul>
           <li>AlibabaCloud AI Search (<code>completion</code>, <code>rerank</code>, <code>sparse_embedding</code>, <code>text_embedding</code>)</li>
           <li>Amazon Bedrock (<code>completion</code>, <code>text_embedding</code>)</li>
+          <li>Amazon SageMaker (<code>chat_completion</code>, <code>completion</code>, <code>rerank</code>, <code>sparse_embedding</code>, <code>text_embedding</code>)</li>
           <li>Anthropic (<code>completion</code>)</li>
           <li>Azure AI Studio (<code>completion</code>, <code>text_embedding</code>)</li>
           <li>Azure OpenAI (<code>completion</code>, <code>text_embedding</code>)</li>
           <li>Cohere (<code>completion</code>, <code>rerank</code>, <code>text_embedding</code>)</li>
+          <li>DeepSeek (<code>chat_completion</code>, <code>completion</code>)</li>
           <li>Elasticsearch (<code>rerank</code>, <code>sparse_embedding</code>, <code>text_embedding</code> - this service is for built-in models and models uploaded through Eland)</li>
           <li>ELSER (<code>sparse_embedding</code>)</li>
           <li>Google AI Studio (<code>completion</code>, <code>text_embedding</code>)</li>
-          <li>Google Vertex AI (<code>rerank</code>, <code>text_embedding</code>)</li>
-          <li>Hugging Face (<code>text_embedding</code>)</li>
-          <li>Mistral (<code>text_embedding</code>)</li>
+          <li>Google Vertex AI (<code>chat_completion</code>, <code>completion</code>, <code>rerank</code>, <code>text_embedding</code>)</li>
+          <li>Hugging Face (<code>chat_completion</code>, <code>completion</code>, <code>rerank</code>, <code>text_embedding</code>)</li>
+          <li>JinaAI (<code>rerank</code>, <code>text_embedding</code>)</li>
+          <li>Llama (<code>chat_completion</code>, <code>completion</code>, <code>text_embedding</code>)</li>
+          <li>Mistral (<code>chat_completion</code>, <code>completion</code>, <code>text_embedding</code>)</li>
           <li>OpenAI (<code>chat_completion</code>, <code>completion</code>, <code>text_embedding</code>)</li>
-          <li>VoyageAI (<code>text_embedding</code>, <code>rerank</code>)</li>
+          <li>VoyageAI (<code>rerank</code>, <code>text_embedding</code>)</li>
           <li>Watsonx inference integration (<code>text_embedding</code>)</li>
-          <li>JinaAI (<code>text_embedding</code>, <code>rerank</code>)</li>
           </ul>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put>`_
 
         :param inference_id: The inference Id
         :param inference_config:
@@ -481,7 +496,7 @@ class InferenceClient(NamespacedClient):
           <p>Create an inference endpoint to perform an inference task with the <code>alibabacloud-ai-search</code> service.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-alibabacloud>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-alibabacloud>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param alibabacloud_inference_id: The unique identifier of the inference endpoint.
@@ -581,7 +596,7 @@ class InferenceClient(NamespacedClient):
           </blockquote>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-amazonbedrock>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-amazonbedrock>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param amazonbedrock_inference_id: The unique identifier of the inference endpoint.
@@ -646,6 +661,112 @@ class InferenceClient(NamespacedClient):
             path_parts=__path_parts,
         )
 
+    @_rewrite_parameters(
+        body_fields=(
+            "service",
+            "service_settings",
+            "chunking_settings",
+            "task_settings",
+        ),
+    )
+    def put_amazonsagemaker(
+        self,
+        *,
+        task_type: t.Union[
+            str,
+            t.Literal[
+                "chat_completion",
+                "completion",
+                "rerank",
+                "sparse_embedding",
+                "text_embedding",
+            ],
+        ],
+        amazonsagemaker_inference_id: str,
+        service: t.Optional[t.Union[str, t.Literal["amazon_sagemaker"]]] = None,
+        service_settings: t.Optional[t.Mapping[str, t.Any]] = None,
+        chunking_settings: t.Optional[t.Mapping[str, t.Any]] = None,
+        error_trace: t.Optional[bool] = None,
+        filter_path: t.Optional[t.Union[str, t.Sequence[str]]] = None,
+        human: t.Optional[bool] = None,
+        pretty: t.Optional[bool] = None,
+        task_settings: t.Optional[t.Mapping[str, t.Any]] = None,
+        timeout: t.Optional[t.Union[str, t.Literal[-1], t.Literal[0]]] = None,
+        body: t.Optional[t.Dict[str, t.Any]] = None,
+    ) -> ObjectApiResponse[t.Any]:
+        """
+        .. raw:: html
+
+          <p>Create an Amazon SageMaker inference endpoint.</p>
+          <p>Create an inference endpoint to perform an inference task with the <code>amazon_sagemaker</code> service.</p>
+
+
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-amazonsagemaker>`_
+
+        :param task_type: The type of the inference task that the model will perform.
+        :param amazonsagemaker_inference_id: The unique identifier of the inference endpoint.
+        :param service: The type of service supported for the specified task type. In
+            this case, `amazon_sagemaker`.
+        :param service_settings: Settings used to install the inference model. These
+            settings are specific to the `amazon_sagemaker` service and `service_settings.api`
+            you specified.
+        :param chunking_settings: The chunking configuration object.
+        :param task_settings: Settings to configure the inference task. These settings
+            are specific to the task type and `service_settings.api` you specified.
+        :param timeout: Specifies the amount of time to wait for the inference endpoint
+            to be created.
+        """
+        if task_type in SKIP_IN_PATH:
+            raise ValueError("Empty value passed for parameter 'task_type'")
+        if amazonsagemaker_inference_id in SKIP_IN_PATH:
+            raise ValueError(
+                "Empty value passed for parameter 'amazonsagemaker_inference_id'"
+            )
+        if service is None and body is None:
+            raise ValueError("Empty value passed for parameter 'service'")
+        if service_settings is None and body is None:
+            raise ValueError("Empty value passed for parameter 'service_settings'")
+        __path_parts: t.Dict[str, str] = {
+            "task_type": _quote(task_type),
+            "amazonsagemaker_inference_id": _quote(amazonsagemaker_inference_id),
+        }
+        __path = f'/_inference/{__path_parts["task_type"]}/{__path_parts["amazonsagemaker_inference_id"]}'
+        __query: t.Dict[str, t.Any] = {}
+        __body: t.Dict[str, t.Any] = body if body is not None else {}
+        if error_trace is not None:
+            __query["error_trace"] = error_trace
+        if filter_path is not None:
+            __query["filter_path"] = filter_path
+        if human is not None:
+            __query["human"] = human
+        if pretty is not None:
+            __query["pretty"] = pretty
+        if timeout is not None:
+            __query["timeout"] = timeout
+        if not __body:
+            if service is not None:
+                __body["service"] = service
+            if service_settings is not None:
+                __body["service_settings"] = service_settings
+            if chunking_settings is not None:
+                __body["chunking_settings"] = chunking_settings
+            if task_settings is not None:
+                __body["task_settings"] = task_settings
+        if not __body:
+            __body = None  # type: ignore[assignment]
+        __headers = {"accept": "application/json"}
+        if __body is not None:
+            __headers["content-type"] = "application/json"
+        return self.perform_request(  # type: ignore[return-value]
+            "PUT",
+            __path,
+            params=__query,
+            headers=__headers,
+            body=__body,
+            endpoint_id="inference.put_amazonsagemaker",
+            path_parts=__path_parts,
+        )
+
     @_rewrite_parameters(
         body_fields=(
             "service",
@@ -677,7 +798,7 @@ class InferenceClient(NamespacedClient):
           <p>Create an inference endpoint to perform an inference task with the <code>anthropic</code> service.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-anthropic>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-anthropic>`_
 
         :param task_type: The task type. The only valid task type for the model to perform
             is `completion`.
@@ -774,7 +895,7 @@ class InferenceClient(NamespacedClient):
           <p>Create an inference endpoint to perform an inference task with the <code>azureaistudio</code> service.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-azureaistudio>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-azureaistudio>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param azureaistudio_inference_id: The unique identifier of the inference endpoint.
@@ -876,7 +997,7 @@ class InferenceClient(NamespacedClient):
           <p>The list of embeddings models that you can choose from in your deployment can be found in the <a href="https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models?tabs=global-standard%2Cstandard-chat-completions#embeddings">Azure models documentation</a>.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-azureopenai>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-azureopenai>`_
 
         :param task_type: The type of the inference task that the model will perform.
             NOTE: The `chat_completion` task type only supports streaming and only through
@@ -974,7 +1095,7 @@ class InferenceClient(NamespacedClient):
           <p>Create an inference endpoint to perform an inference task with the <code>cohere</code> service.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-cohere>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-cohere>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param cohere_inference_id: The unique identifier of the inference endpoint.
@@ -1037,6 +1158,221 @@ class InferenceClient(NamespacedClient):
             path_parts=__path_parts,
         )
 
+    @_rewrite_parameters(
+        body_fields=(
+            "service",
+            "service_settings",
+            "chunking_settings",
+            "task_settings",
+        ),
+    )
+    def put_custom(
+        self,
+        *,
+        task_type: t.Union[
+            str, t.Literal["completion", "rerank", "sparse_embedding", "text_embedding"]
+        ],
+        custom_inference_id: str,
+        service: t.Optional[t.Union[str, t.Literal["custom"]]] = None,
+        service_settings: t.Optional[t.Mapping[str, t.Any]] = None,
+        chunking_settings: t.Optional[t.Mapping[str, t.Any]] = None,
+        error_trace: t.Optional[bool] = None,
+        filter_path: t.Optional[t.Union[str, t.Sequence[str]]] = None,
+        human: t.Optional[bool] = None,
+        pretty: t.Optional[bool] = None,
+        task_settings: t.Optional[t.Mapping[str, t.Any]] = None,
+        body: t.Optional[t.Dict[str, t.Any]] = None,
+    ) -> ObjectApiResponse[t.Any]:
+        """
+        .. raw:: html
+
+          <p>Create a custom inference endpoint.</p>
+          <p>The custom service gives more control over how to interact with external inference services that aren't explicitly supported through dedicated integrations.
+          The custom service gives you the ability to define the headers, url, query parameters, request body, and secrets.
+          The custom service supports the template replacement functionality, which enables you to define a template that can be replaced with the value associated with that key.
+          Templates are portions of a string that start with <code>${</code> and end with <code>}</code>.
+          The parameters <code>secret_parameters</code> and <code>task_settings</code> are checked for keys for template replacement. Template replacement is supported in the <code>request</code>, <code>headers</code>, <code>url</code>, and <code>query_parameters</code>.
+          If the definition (key) is not found for a template, an error message is returned.
+          In case of an endpoint definition like the following:</p>
+          <pre><code>PUT _inference/text_embedding/test-text-embedding
+          {
+            &quot;service&quot;: &quot;custom&quot;,
+            &quot;service_settings&quot;: {
+               &quot;secret_parameters&quot;: {
+                    &quot;api_key&quot;: &quot;&lt;some api key&gt;&quot;
+               },
+               &quot;url&quot;: &quot;...endpoints.huggingface.cloud/v1/embeddings&quot;,
+               &quot;headers&quot;: {
+                   &quot;Authorization&quot;: &quot;Bearer ${api_key}&quot;,
+                   &quot;Content-Type&quot;: &quot;application/json&quot;
+               },
+               &quot;request&quot;: &quot;{\\&quot;input\\&quot;: ${input}}&quot;,
+               &quot;response&quot;: {
+                   &quot;json_parser&quot;: {
+                       &quot;text_embeddings&quot;:&quot;$.data[*].embedding[*]&quot;
+                   }
+               }
+            }
+          }
+          </code></pre>
+          <p>To replace <code>${api_key}</code> the <code>secret_parameters</code> and <code>task_settings</code> are checked for a key named <code>api_key</code>.</p>
+          <blockquote>
+          <p>info
+          Templates should not be surrounded by quotes.</p>
+          </blockquote>
+          <p>Pre-defined templates:</p>
+          <ul>
+          <li><code>${input}</code> refers to the array of input strings that comes from the <code>input</code> field of the subsequent inference requests.</li>
+          <li><code>${input_type}</code> refers to the input type translation values.</li>
+          <li><code>${query}</code> refers to the query field used specifically for reranking tasks.</li>
+          <li><code>${top_n}</code> refers to the <code>top_n</code> field available when performing rerank requests.</li>
+          <li><code>${return_documents}</code> refers to the <code>return_documents</code> field available when performing rerank requests.</li>
+          </ul>
+
+
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-custom>`_
+
+        :param task_type: The type of the inference task that the model will perform.
+        :param custom_inference_id: The unique identifier of the inference endpoint.
+        :param service: The type of service supported for the specified task type. In
+            this case, `custom`.
+        :param service_settings: Settings used to install the inference model. These
+            settings are specific to the `custom` service.
+        :param chunking_settings: The chunking configuration object.
+        :param task_settings: Settings to configure the inference task. These settings
+            are specific to the task type you specified.
+        """
+        if task_type in SKIP_IN_PATH:
+            raise ValueError("Empty value passed for parameter 'task_type'")
+        if custom_inference_id in SKIP_IN_PATH:
+            raise ValueError("Empty value passed for parameter 'custom_inference_id'")
+        if service is None and body is None:
+            raise ValueError("Empty value passed for parameter 'service'")
+        if service_settings is None and body is None:
+            raise ValueError("Empty value passed for parameter 'service_settings'")
+        __path_parts: t.Dict[str, str] = {
+            "task_type": _quote(task_type),
+            "custom_inference_id": _quote(custom_inference_id),
+        }
+        __path = f'/_inference/{__path_parts["task_type"]}/{__path_parts["custom_inference_id"]}'
+        __query: t.Dict[str, t.Any] = {}
+        __body: t.Dict[str, t.Any] = body if body is not None else {}
+        if error_trace is not None:
+            __query["error_trace"] = error_trace
+        if filter_path is not None:
+            __query["filter_path"] = filter_path
+        if human is not None:
+            __query["human"] = human
+        if pretty is not None:
+            __query["pretty"] = pretty
+        if not __body:
+            if service is not None:
+                __body["service"] = service
+            if service_settings is not None:
+                __body["service_settings"] = service_settings
+            if chunking_settings is not None:
+                __body["chunking_settings"] = chunking_settings
+            if task_settings is not None:
+                __body["task_settings"] = task_settings
+        if not __body:
+            __body = None  # type: ignore[assignment]
+        __headers = {"accept": "application/json"}
+        if __body is not None:
+            __headers["content-type"] = "application/json"
+        return self.perform_request(  # type: ignore[return-value]
+            "PUT",
+            __path,
+            params=__query,
+            headers=__headers,
+            body=__body,
+            endpoint_id="inference.put_custom",
+            path_parts=__path_parts,
+        )
+
+    @_rewrite_parameters(
+        body_fields=("service", "service_settings", "chunking_settings"),
+    )
+    def put_deepseek(
+        self,
+        *,
+        task_type: t.Union[str, t.Literal["chat_completion", "completion"]],
+        deepseek_inference_id: str,
+        service: t.Optional[t.Union[str, t.Literal["deepseek"]]] = None,
+        service_settings: t.Optional[t.Mapping[str, t.Any]] = None,
+        chunking_settings: t.Optional[t.Mapping[str, t.Any]] = None,
+        error_trace: t.Optional[bool] = None,
+        filter_path: t.Optional[t.Union[str, t.Sequence[str]]] = None,
+        human: t.Optional[bool] = None,
+        pretty: t.Optional[bool] = None,
+        timeout: t.Optional[t.Union[str, t.Literal[-1], t.Literal[0]]] = None,
+        body: t.Optional[t.Dict[str, t.Any]] = None,
+    ) -> ObjectApiResponse[t.Any]:
+        """
+        .. raw:: html
+
+          <p>Create a DeepSeek inference endpoint.</p>
+          <p>Create an inference endpoint to perform an inference task with the <code>deepseek</code> service.</p>
+
+
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-deepseek>`_
+
+        :param task_type: The type of the inference task that the model will perform.
+        :param deepseek_inference_id: The unique identifier of the inference endpoint.
+        :param service: The type of service supported for the specified task type. In
+            this case, `deepseek`.
+        :param service_settings: Settings used to install the inference model. These
+            settings are specific to the `deepseek` service.
+        :param chunking_settings: The chunking configuration object.
+        :param timeout: Specifies the amount of time to wait for the inference endpoint
+            to be created.
+        """
+        if task_type in SKIP_IN_PATH:
+            raise ValueError("Empty value passed for parameter 'task_type'")
+        if deepseek_inference_id in SKIP_IN_PATH:
+            raise ValueError("Empty value passed for parameter 'deepseek_inference_id'")
+        if service is None and body is None:
+            raise ValueError("Empty value passed for parameter 'service'")
+        if service_settings is None and body is None:
+            raise ValueError("Empty value passed for parameter 'service_settings'")
+        __path_parts: t.Dict[str, str] = {
+            "task_type": _quote(task_type),
+            "deepseek_inference_id": _quote(deepseek_inference_id),
+        }
+        __path = f'/_inference/{__path_parts["task_type"]}/{__path_parts["deepseek_inference_id"]}'
+        __query: t.Dict[str, t.Any] = {}
+        __body: t.Dict[str, t.Any] = body if body is not None else {}
+        if error_trace is not None:
+            __query["error_trace"] = error_trace
+        if filter_path is not None:
+            __query["filter_path"] = filter_path
+        if human is not None:
+            __query["human"] = human
+        if pretty is not None:
+            __query["pretty"] = pretty
+        if timeout is not None:
+            __query["timeout"] = timeout
+        if not __body:
+            if service is not None:
+                __body["service"] = service
+            if service_settings is not None:
+                __body["service_settings"] = service_settings
+            if chunking_settings is not None:
+                __body["chunking_settings"] = chunking_settings
+        if not __body:
+            __body = None  # type: ignore[assignment]
+        __headers = {"accept": "application/json"}
+        if __body is not None:
+            __headers["content-type"] = "application/json"
+        return self.perform_request(  # type: ignore[return-value]
+            "PUT",
+            __path,
+            params=__query,
+            headers=__headers,
+            body=__body,
+            endpoint_id="inference.put_deepseek",
+            path_parts=__path_parts,
+        )
+
     @_rewrite_parameters(
         body_fields=(
             "service",
@@ -1083,7 +1419,7 @@ class InferenceClient(NamespacedClient):
           Avoid creating multiple endpoints for the same model unless required, as each endpoint consumes significant resources.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-elasticsearch>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-elasticsearch>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param elasticsearch_inference_id: The unique identifier of the inference endpoint.
@@ -1188,7 +1524,7 @@ class InferenceClient(NamespacedClient):
           Avoid creating multiple endpoints for the same model unless required, as each endpoint consumes significant resources.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-elser>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-elser>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param elser_inference_id: The unique identifier of the inference endpoint.
@@ -1272,7 +1608,7 @@ class InferenceClient(NamespacedClient):
           <p>Create an inference endpoint to perform an inference task with the <code>googleaistudio</code> service.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-googleaistudio>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-googleaistudio>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param googleaistudio_inference_id: The unique identifier of the inference endpoint.
@@ -1344,7 +1680,9 @@ class InferenceClient(NamespacedClient):
     def put_googlevertexai(
         self,
         *,
-        task_type: t.Union[str, t.Literal["rerank", "text_embedding"]],
+        task_type: t.Union[
+            str, t.Literal["chat_completion", "completion", "rerank", "text_embedding"]
+        ],
         googlevertexai_inference_id: str,
         service: t.Optional[t.Union[str, t.Literal["googlevertexai"]]] = None,
         service_settings: t.Optional[t.Mapping[str, t.Any]] = None,
@@ -1364,7 +1702,7 @@ class InferenceClient(NamespacedClient):
           <p>Create an inference endpoint to perform an inference task with the <code>googlevertexai</code> service.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-googlevertexai>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-googlevertexai>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param googlevertexai_inference_id: The unique identifier of the inference endpoint.
@@ -1430,12 +1768,19 @@ class InferenceClient(NamespacedClient):
         )
 
     @_rewrite_parameters(
-        body_fields=("service", "service_settings", "chunking_settings"),
+        body_fields=(
+            "service",
+            "service_settings",
+            "chunking_settings",
+            "task_settings",
+        ),
     )
     def put_hugging_face(
         self,
         *,
-        task_type: t.Union[str, t.Literal["text_embedding"]],
+        task_type: t.Union[
+            str, t.Literal["chat_completion", "completion", "rerank", "text_embedding"]
+        ],
         huggingface_inference_id: str,
         service: t.Optional[t.Union[str, t.Literal["hugging_face"]]] = None,
         service_settings: t.Optional[t.Mapping[str, t.Any]] = None,
@@ -1444,6 +1789,7 @@ class InferenceClient(NamespacedClient):
         filter_path: t.Optional[t.Union[str, t.Sequence[str]]] = None,
         human: t.Optional[bool] = None,
         pretty: t.Optional[bool] = None,
+        task_settings: t.Optional[t.Mapping[str, t.Any]] = None,
         timeout: t.Optional[t.Union[str, t.Literal[-1], t.Literal[0]]] = None,
         body: t.Optional[t.Dict[str, t.Any]] = None,
     ) -> ObjectApiResponse[t.Any]:
@@ -1451,11 +1797,14 @@ class InferenceClient(NamespacedClient):
         .. raw:: html
 
           <p>Create a Hugging Face inference endpoint.</p>
-          <p>Create an inference endpoint to perform an inference task with the <code>hugging_face</code> service.</p>
-          <p>You must first create an inference endpoint on the Hugging Face endpoint page to get an endpoint URL.
-          Select the model you want to use on the new endpoint creation page (for example <code>intfloat/e5-small-v2</code>), then select the sentence embeddings task under the advanced configuration section.
-          Create the endpoint and copy the URL after the endpoint initialization has been finished.</p>
-          <p>The following models are recommended for the Hugging Face service:</p>
+          <p>Create an inference endpoint to perform an inference task with the <code>hugging_face</code> service.
+          Supported tasks include: <code>text_embedding</code>, <code>completion</code>, and <code>chat_completion</code>.</p>
+          <p>To configure the endpoint, first visit the Hugging Face Inference Endpoints page and create a new endpoint.
+          Select a model that supports the task you intend to use.</p>
+          <p>For Elastic's <code>text_embedding</code> task:
+          The selected model must support the <code>Sentence Embeddings</code> task. On the new endpoint creation page, select the <code>Sentence Embeddings</code> task under the <code>Advanced Configuration</code> section.
+          After the endpoint has initialized, copy the generated endpoint URL.
+          Recommended models for <code>text_embedding</code> task:</p>
           <ul>
           <li><code>all-MiniLM-L6-v2</code></li>
           <li><code>all-MiniLM-L12-v2</code></li>
@@ -1465,9 +1814,27 @@ class InferenceClient(NamespacedClient):
           <li><code>multilingual-e5-base</code></li>
           <li><code>multilingual-e5-small</code></li>
           </ul>
+          <p>For Elastic's <code>chat_completion</code> and <code>completion</code> tasks:
+          The selected model must support the <code>Text Generation</code> task and expose OpenAI API. HuggingFace supports both serverless and dedicated endpoints for <code>Text Generation</code>. When creating dedicated endpoint select the <code>Text Generation</code> task.
+          After the endpoint is initialized (for dedicated) or ready (for serverless), ensure it supports the OpenAI API and includes <code>/v1/chat/completions</code> part in URL. Then, copy the full endpoint URL for use.
+          Recommended models for <code>chat_completion</code> and <code>completion</code> tasks:</p>
+          <ul>
+          <li><code>Mistral-7B-Instruct-v0.2</code></li>
+          <li><code>QwQ-32B</code></li>
+          <li><code>Phi-3-mini-128k-instruct</code></li>
+          </ul>
+          <p>For Elastic's <code>rerank</code> task:
+          The selected model must support the <code>sentence-ranking</code> task and expose OpenAI API.
+          HuggingFace supports only dedicated (not serverless) endpoints for <code>Rerank</code> so far.
+          After the endpoint is initialized, copy the full endpoint URL for use.
+          Tested models for <code>rerank</code> task:</p>
+          <ul>
+          <li><code>bge-reranker-base</code></li>
+          <li><code>jina-reranker-v1-turbo-en-GGUF</code></li>
+          </ul>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-hugging-face>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-hugging-face>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param huggingface_inference_id: The unique identifier of the inference endpoint.
@@ -1476,6 +1843,8 @@ class InferenceClient(NamespacedClient):
         :param service_settings: Settings used to install the inference model. These
             settings are specific to the `hugging_face` service.
         :param chunking_settings: The chunking configuration object.
+        :param task_settings: Settings to configure the inference task. These settings
+            are specific to the task type you specified.
         :param timeout: Specifies the amount of time to wait for the inference endpoint
             to be created.
         """
@@ -1513,6 +1882,8 @@ class InferenceClient(NamespacedClient):
                 __body["service_settings"] = service_settings
             if chunking_settings is not None:
                 __body["chunking_settings"] = chunking_settings
+            if task_settings is not None:
+                __body["task_settings"] = task_settings
         if not __body:
             __body = None  # type: ignore[assignment]
         __headers = {"accept": "application/json"}
@@ -1561,7 +1932,7 @@ class InferenceClient(NamespacedClient):
           To review the available <code>text_embedding</code> models, refer to the <a href="https://jina.ai/embeddings/">https://jina.ai/embeddings/</a>.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-jinaai>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-jinaai>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param jinaai_inference_id: The unique identifier of the inference endpoint.
@@ -1630,7 +2001,9 @@ class InferenceClient(NamespacedClient):
     def put_mistral(
         self,
         *,
-        task_type: t.Union[str, t.Literal["text_embedding"]],
+        task_type: t.Union[
+            str, t.Literal["chat_completion", "completion", "text_embedding"]
+        ],
         mistral_inference_id: str,
         service: t.Optional[t.Union[str, t.Literal["mistral"]]] = None,
         service_settings: t.Optional[t.Mapping[str, t.Any]] = None,
@@ -1646,13 +2019,12 @@ class InferenceClient(NamespacedClient):
         .. raw:: html
 
           <p>Create a Mistral inference endpoint.</p>
-          <p>Creates an inference endpoint to perform an inference task with the <code>mistral</code> service.</p>
+          <p>Create an inference endpoint to perform an inference task with the <code>mistral</code> service.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-mistral>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-mistral>`_
 
-        :param task_type: The task type. The only valid task type for the model to perform
-            is `text_embedding`.
+        :param task_type: The type of the inference task that the model will perform.
         :param mistral_inference_id: The unique identifier of the inference endpoint.
         :param service: The type of service supported for the specified task type. In
             this case, `mistral`.
@@ -1742,7 +2114,7 @@ class InferenceClient(NamespacedClient):
           <p>Create an inference endpoint to perform an inference task with the <code>openai</code> service or <code>openai</code> compatible APIs.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-openai>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-openai>`_
 
         :param task_type: The type of the inference task that the model will perform.
             NOTE: The `chat_completion` task type only supports streaming and only through
@@ -1839,7 +2211,7 @@ class InferenceClient(NamespacedClient):
           <p>Avoid creating multiple endpoints for the same model unless required, as each endpoint consumes significant resources.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-voyageai>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-voyageai>`_
 
         :param task_type: The type of the inference task that the model will perform.
         :param voyageai_inference_id: The unique identifier of the inference endpoint.
@@ -1908,7 +2280,9 @@ class InferenceClient(NamespacedClient):
     def put_watsonx(
         self,
         *,
-        task_type: t.Union[str, t.Literal["text_embedding"]],
+        task_type: t.Union[
+            str, t.Literal["chat_completion", "completion", "text_embedding"]
+        ],
         watsonx_inference_id: str,
         service: t.Optional[t.Union[str, t.Literal["watsonxai"]]] = None,
         service_settings: t.Optional[t.Mapping[str, t.Any]] = None,
@@ -1928,10 +2302,9 @@ class InferenceClient(NamespacedClient):
           You can provision one through the IBM catalog, the Cloud Databases CLI plug-in, the Cloud Databases API, or Terraform.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-put-watsonx>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-watsonx>`_
 
-        :param task_type: The task type. The only valid task type for the model to perform
-            is `text_embedding`.
+        :param task_type: The type of the inference task that the model will perform.
         :param watsonx_inference_id: The unique identifier of the inference endpoint.
         :param service: The type of service supported for the specified task type. In
             this case, `watsonxai`.
@@ -2008,7 +2381,7 @@ class InferenceClient(NamespacedClient):
           <p>Perform reranking inference on the service</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-inference>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-inference>`_
 
         :param inference_id: The unique identifier for the inference endpoint.
         :param input: The text on which you want to perform the inference task. It can
@@ -2084,7 +2457,7 @@ class InferenceClient(NamespacedClient):
           <p>Perform sparse embedding inference on the service</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-inference>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-inference>`_
 
         :param inference_id: The inference Id
         :param input: Inference input. Either a string or an array of strings.
@@ -2152,7 +2525,7 @@ class InferenceClient(NamespacedClient):
           <p>Perform text embedding inference on the service</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-inference>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-inference>`_
 
         :param inference_id: The inference Id
         :param input: Inference input. Either a string or an array of strings.
@@ -2234,7 +2607,7 @@ class InferenceClient(NamespacedClient):
           However, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.</p>
 
 
-        `<https://www.elastic.co/docs/api/doc/elasticsearch/v9/operation/operation-inference-update>`_
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-update>`_
 
         :param inference_id: The unique identifier of the inference endpoint.
         :param inference_config: