elasticsearch 9.1.1__py3-none-any.whl → 9.2.0__py3-none-any.whl

elasticsearch/_sync/client/inference.py

@@ -78,11 +78,7 @@ class InferenceClient(NamespacedClient):
                 __body["input"] = input
             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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "POST",
             __path,
@@ -338,11 +334,7 @@ class InferenceClient(NamespacedClient):
                 __body["query"] = query
             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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "POST",
             __path,
@@ -389,11 +381,12 @@ 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>
           <p>The following integrations are available through the inference API. You can find the available task types next to the integration name:</p>
           <ul>
+          <li>AI21 (<code>chat_completion</code>, <code>completion</code>)</li>
           <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 AI Studio (<code>completion</code>, 'rerank', <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>
@@ -463,6 +456,82 @@ class InferenceClient(NamespacedClient):
             path_parts=__path_parts,
         )
 
+    @_rewrite_parameters(
+        body_fields=("service", "service_settings"),
+    )
+    def put_ai21(
+        self,
+        *,
+        task_type: t.Union[str, t.Literal["chat_completion", "completion"]],
+        ai21_inference_id: str,
+        service: t.Optional[t.Union[str, t.Literal["ai21"]]] = None,
+        service_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 AI21 inference endpoint.</p>
+          <p>Create an inference endpoint to perform an inference task with the <code>ai21</code> service.</p>
+
+
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-ai21>`_
+
+        :param task_type: The type of the inference task that the model will perform.
+        :param ai21_inference_id: The unique identifier of the inference endpoint.
+        :param service: The type of service supported for the specified task type. In
+            this case, `ai21`.
+        :param service_settings: Settings used to install the inference model. These
+            settings are specific to the `ai21` service.
+        :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 ai21_inference_id in SKIP_IN_PATH:
+            raise ValueError("Empty value passed for parameter 'ai21_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),
+            "ai21_inference_id": _quote(ai21_inference_id),
+        }
+        __path = f'/_inference/{__path_parts["task_type"]}/{__path_parts["ai21_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
+        __headers = {"accept": "application/json", "content-type": "application/json"}
+        return self.perform_request(  # type: ignore[return-value]
+            "PUT",
+            __path,
+            params=__query,
+            headers=__headers,
+            body=__body,
+            endpoint_id="inference.put_ai21",
+            path_parts=__path_parts,
+        )
+
     @_rewrite_parameters(
         body_fields=(
             "service",
@@ -546,11 +615,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -646,11 +711,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -752,11 +813,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -849,11 +906,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -875,7 +928,7 @@ class InferenceClient(NamespacedClient):
     def put_azureaistudio(
         self,
         *,
-        task_type: t.Union[str, t.Literal["completion", "text_embedding"]],
+        task_type: t.Union[str, t.Literal["completion", "rerank", "text_embedding"]],
         azureaistudio_inference_id: str,
         service: t.Optional[t.Union[str, t.Literal["azureaistudio"]]] = None,
         service_settings: t.Optional[t.Mapping[str, t.Any]] = None,
@@ -945,11 +998,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1049,11 +1098,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1143,11 +1188,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1158,6 +1199,99 @@ class InferenceClient(NamespacedClient):
             path_parts=__path_parts,
         )
 
+    @_rewrite_parameters(
+        body_fields=(
+            "service",
+            "service_settings",
+            "chunking_settings",
+            "task_settings",
+        ),
+    )
+    def put_contextualai(
+        self,
+        *,
+        task_type: t.Union[str, t.Literal["rerank"]],
+        contextualai_inference_id: str,
+        service: t.Optional[t.Union[str, t.Literal["contextualai"]]] = 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 Contextual AI inference endpoint.</p>
+          <p>Create an inference endpoint to perform an inference task with the <code>contexualai</code> service.</p>
+          <p>To review the available <code>rerank</code> models, refer to <a href="https://docs.contextual.ai/api-reference/rerank/rerank#body-model">https://docs.contextual.ai/api-reference/rerank/rerank#body-model</a>.</p>
+
+
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-contextualai>`_
+
+        :param task_type: The type of the inference task that the model will perform.
+        :param contextualai_inference_id: The unique identifier of the inference endpoint.
+        :param service: The type of service supported for the specified task type. In
+            this case, `contextualai`.
+        :param service_settings: Settings used to install the inference model. These
+            settings are specific to the `contextualai` 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.
+        """
+        if task_type in SKIP_IN_PATH:
+            raise ValueError("Empty value passed for parameter 'task_type'")
+        if contextualai_inference_id in SKIP_IN_PATH:
+            raise ValueError(
+                "Empty value passed for parameter 'contextualai_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),
+            "contextualai_inference_id": _quote(contextualai_inference_id),
+        }
+        __path = f'/_inference/{__path_parts["task_type"]}/{__path_parts["contextualai_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
+        __headers = {"accept": "application/json", "content-type": "application/json"}
+        return self.perform_request(  # type: ignore[return-value]
+            "PUT",
+            __path,
+            params=__query,
+            headers=__headers,
+            body=__body,
+            endpoint_id="inference.put_contextualai",
+            path_parts=__path_parts,
+        )
+
     @_rewrite_parameters(
         body_fields=(
             "service",
@@ -1274,11 +1408,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1358,11 +1488,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1470,11 +1596,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1532,7 +1654,8 @@ class InferenceClient(NamespacedClient):
             this case, `elser`.
         :param service_settings: Settings used to install the inference model. These
             settings are specific to the `elser` service.
-        :param chunking_settings: The chunking configuration object.
+        :param chunking_settings: The chunking configuration object. Note that for ELSER
+            endpoints, the max_chunk_size may not exceed `300`.
         :param timeout: Specifies the amount of time to wait for the inference endpoint
             to be created.
         """
@@ -1568,11 +1691,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1654,11 +1773,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1752,11 +1867,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1884,11 +1995,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1980,11 +2087,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -1995,6 +2098,88 @@ class InferenceClient(NamespacedClient):
             path_parts=__path_parts,
         )
 
+    @_rewrite_parameters(
+        body_fields=("service", "service_settings", "chunking_settings"),
+    )
+    def put_llama(
+        self,
+        *,
+        task_type: t.Union[
+            str, t.Literal["chat_completion", "completion", "text_embedding"]
+        ],
+        llama_inference_id: str,
+        service: t.Optional[t.Union[str, t.Literal["llama"]]] = 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 Llama inference endpoint.</p>
+          <p>Create an inference endpoint to perform an inference task with the <code>llama</code> service.</p>
+
+
+        `<https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-llama>`_
+
+        :param task_type: The type of the inference task that the model will perform.
+        :param llama_inference_id: The unique identifier of the inference endpoint.
+        :param service: The type of service supported for the specified task type. In
+            this case, `llama`.
+        :param service_settings: Settings used to install the inference model. These
+            settings are specific to the `llama` 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 llama_inference_id in SKIP_IN_PATH:
+            raise ValueError("Empty value passed for parameter 'llama_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),
+            "llama_inference_id": _quote(llama_inference_id),
+        }
+        __path = f'/_inference/{__path_parts["task_type"]}/{__path_parts["llama_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
+        __headers = {"accept": "application/json", "content-type": "application/json"}
+        return self.perform_request(  # type: ignore[return-value]
+            "PUT",
+            __path,
+            params=__query,
+            headers=__headers,
+            body=__body,
+            endpoint_id="inference.put_llama",
+            path_parts=__path_parts,
+        )
+
     @_rewrite_parameters(
         body_fields=("service", "service_settings", "chunking_settings"),
     )
@@ -2066,11 +2251,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -2164,11 +2345,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -2259,11 +2436,7 @@ class InferenceClient(NamespacedClient):
                 __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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -2343,11 +2516,7 @@ class InferenceClient(NamespacedClient):
                 __body["service"] = service
             if service_settings is not None:
                 __body["service_settings"] = service_settings
-        if not __body:
-            __body = None  # type: ignore[assignment]
-        __headers = {"accept": "application/json"}
-        if __body is not None:
-            __headers["content-type"] = "application/json"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "PUT",
             __path,
@@ -2420,11 +2589,7 @@ class InferenceClient(NamespacedClient):
                 __body["query"] = query
             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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "POST",
             __path,
@@ -2488,11 +2653,7 @@ class InferenceClient(NamespacedClient):
                 __body["input"] = input
             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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "POST",
             __path,
@@ -2504,7 +2665,7 @@ class InferenceClient(NamespacedClient):
         )
 
     @_rewrite_parameters(
-        body_fields=("input", "task_settings"),
+        body_fields=("input", "input_type", "task_settings"),
     )
     def text_embedding(
         self,
@@ -2514,6 +2675,7 @@ 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,
         task_settings: t.Optional[t.Any] = None,
         timeout: t.Optional[t.Union[str, t.Literal[-1], t.Literal[0]]] = None,
@@ -2529,6 +2691,13 @@ class InferenceClient(NamespacedClient):
 
         :param inference_id: The inference Id
         :param input: Inference input. Either a string or an array of strings.
+        :param input_type: The input data type for the text embedding model. 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 task_settings: Optional task settings
         :param timeout: Specifies the amount of time to wait for the inference request
             to complete.
@@ -2554,13 +2723,11 @@ 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 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"
+        __headers = {"accept": "application/json", "content-type": "application/json"}
         return self.perform_request(  # type: ignore[return-value]
             "POST",
             __path,