[ML] Add DeepSeek

specification/_doc_ids/table.csv

@@ -165,6 +165,7 @@ data-stream-update,https://www.elastic.co/docs/api/doc/elasticsearch/operation/o
 data-streams,https://www.elastic.co/docs/manage-data/data-store/data-streams,
 date-index-name-processor,https://www.elastic.co/docs/reference/enrich-processor/date-index-name-processor,
 dcg,https://www.elastic.co/docs/reference/elasticsearch/rest-apis/search-rank-eval#_discounted_cumulative_gain_dcg,
+deepseek-api-keys,https://api-docs.deepseek.com/,
 defining-roles,https://www.elastic.co/docs/deploy-manage/users-roles/cluster-or-deployment-auth/defining-roles,
 delete-analytics-collection,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-search-application-delete-behavioral-analytics,https://www.elastic.co/guide/en/elasticsearch/reference/8.18/delete-analytics-collection.html
 delete-async-sql-search-api,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-sql-delete-async,https://www.elastic.co/guide/en/elasticsearch/reference/8.18/delete-async-sql-search-api.html
@@ -354,6 +355,7 @@ inference-api-put-anthropic,https://www.elastic.co/docs/api/doc/elasticsearch/op
 inference-api-put-azureaistudio,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-azureaistudio,https://www.elastic.co/guide/en/elasticsearch/reference/8.18/infer-service-azure-ai-studio.html
 inference-api-put-azureopenai,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-azureopenai,https://www.elastic.co/guide/en/elasticsearch/reference/8.18/infer-service-azure-openai.html
 inference-api-put-cohere,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-cohere,https://www.elastic.co/guide/en/elasticsearch/reference/8.18/infer-service-cohere.html
+inference-api-put-deepseek,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-deepseek,https://www.elastic.co/guide/en/elasticsearch/reference/8.18/infer-service-deepseek.html
 inference-api-put-eis,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-eis,
 inference-api-put-elasticsearch,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-elasticsearch,https://www.elastic.co/guide/en/elasticsearch/reference/8.18/infer-service-elasticsearch.html
 inference-api-put-elser,https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation-inference-put-elser,https://www.elastic.co/guide/en/elasticsearch/reference/8.18/infer-service-elser.html
@@ -924,4 +926,4 @@ watcher-api-update-settings,https://www.elastic.co/docs/api/doc/elasticsearch/op
 watsonx-api-keys,https://cloud.ibm.com/iam/apikeys,
 watsonx-api-models,https://www.ibm.com/products/watsonx-ai/foundation-models,
 watsonx-api-version,https://cloud.ibm.com/apidocs/watsonx-ai#active-version-dates,
-xpack-rollup,https://www.elastic.co/docs/manage-data/lifecycle/rollup,
+xpack-rollup,https://www.elastic.co/docs/manage-data/lifecycle/rollup,

specification/inference/_types/CommonTypes.ts

@@ -770,6 +770,39 @@ export class EisServiceSettings {
   rate_limit?: RateLimitSetting
 }
 
+export class DeepSeekServiceSettings {
+  /**
+   * A valid API key for your DeepSeek account.
+   * You can find or create your DeepSeek API keys on the DeepSeek API key page.
+   *
+   * IMPORTANT: You need to provide the API key only once, during the inference model creation.
+   * The get inference endpoint API does not retrieve your API key.
+   * After creating the inference model, you cannot change the associated API key.
+   * If you want to use a different API key, delete the inference model and recreate it with the same name and the updated API key.
+   * @ext_doc_id deepseek-api-keys
+   */
+  api_key: string
+  /**
+   * For a `completion` or `chat_completion` task, the name of the model to use for the inference task.
+   *
+   * * For the available `completion` and `chat_completion` models, refer to the [DeepSeek Models & Pricing docs](https://api-docs.deepseek.com/quick_start/pricing).
+   */
+  model_id: string
+  /**
+   * The URL endpoint to use for the requests. Defaults to `https://api.deepseek.com/chat/completions`.
+   */
+  url?: string
+}
+
+export enum DeepSeekTaskType {
+  completion,
+  chat_completion
+}
+
+export enum DeepSeekServiceType {
+  deepseek
+}
+
 export enum EisTaskType {
   chat_completion
 }

specification/inference/_types/Services.ts

@@ -27,6 +27,7 @@ import {
   TaskTypeAzureAIStudio,
   TaskTypeAzureOpenAI,
   TaskTypeCohere,
+  TaskTypeDeepSeek,
   TaskTypeElasticsearch,
   TaskTypeELSER,
   TaskTypeGoogleAIStudio,
@@ -152,6 +153,17 @@ export class InferenceEndpointInfoCohere extends InferenceEndpoint {
   task_type: TaskTypeCohere
 }
 
+export class InferenceEndpointInfoDeepSeek extends InferenceEndpoint {
+  /**
+   * The inference Id
+   */
+  inference_id: string
+  /**
+   * The task type
+   */
+  task_type: TaskTypeDeepSeek
+}
+
 export class InferenceEndpointInfoElasticsearch extends InferenceEndpoint {
   /**
    * The inference Id

specification/inference/_types/TaskType.ts

@@ -65,6 +65,11 @@ export enum TaskTypeCohere {
   completion
 }
 
+export enum TaskTypeDeepSeek {
+  completion,
+  chat_completion
+}
+
 export enum TaskTypeElasticsearch {
   sparse_embedding,
   text_embedding,

specification/inference/put/PutRequest.ts

@@ -36,6 +36,7 @@ import { TaskType } from '@inference/_types/TaskType'
  * * Azure AI Studio (`completion`, `text_embedding`)
  * * Azure OpenAI (`completion`, `text_embedding`)
  * * Cohere (`completion`, `rerank`, `text_embedding`)
+ * * DeepSeek (`completion`, `chat_completion`)
  * * Elasticsearch (`rerank`, `sparse_embedding`, `text_embedding` - this service is for built-in models and models uploaded through Eland)
  * * ELSER (`sparse_embedding`)
  * * Google AI Studio (`completion`, `text_embedding`)

specification/inference/put_deepseek/PutDeepSeekRequest.ts

@@ -0,0 +1,72 @@
+/*
+ * Licensed to Elasticsearch B.V. under one or more contributor
+ * license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright
+ * ownership. Elasticsearch B.V. licenses this file to you under
+ * the Apache License, Version 2.0 (the "License"); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied.  See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+import { RequestBase } from '@_types/Base'
+import { Id } from '@_types/common'
+import {
+  DeepSeekServiceSettings,
+  DeepSeekServiceType,
+  DeepSeekTaskType
+} from '@inference/_types/CommonTypes'
+import { InferenceChunkingSettings } from '@inference/_types/Services'
+
+/**
+ * Create a DeepSeek inference endpoint.
+ *
+ * Create an inference endpoint to perform an inference task with the `deepseek` service.
+ * @rest_spec_name inference.put_deepseek
+ * @availability stack since=9.1.0 stability=stable visibility=public
+ * @availability serverless stability=stable visibility=public
+ * @cluster_privileges manage_inference
+ * @doc_id inference-api-put-deepseek
+ */
+export interface Request extends RequestBase {
+  urls: [
+    {
+      path: '/_inference/{task_type}/{deepseek_inference_id}'
+      methods: ['PUT']
+    }
+  ]
+  path_parts: {
+    /**
+     * The type of the inference task that the model will perform.
+     */
+    task_type: DeepSeekTaskType
+    /**
+     * The unique identifier of the inference endpoint.
+     */
+    deepseek_inference_id: Id
+  }
+  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`.
+     */
+    service: DeepSeekServiceType
+    /**
+     * Settings used to install the inference model.
+     * These settings are specific to the `deepseek` service.
+     */
+    service_settings: DeepSeekServiceSettings
+  }
+}

specification/inference/put_deepseek/PutDeepSeekResponse.ts

@@ -0,0 +1,25 @@
+/*
+ * Licensed to Elasticsearch B.V. under one or more contributor
+ * license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright
+ * ownership. Elasticsearch B.V. licenses this file to you under
+ * the Apache License, Version 2.0 (the "License"); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied.  See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+import { InferenceEndpointInfoDeepSeek } from '@inference/_types/Services'
+
+export class Response {
+  /** @codegen_name endpoint_info */
+  body: InferenceEndpointInfoDeepSeek
+}

specification/inference/put_deepseek/request/PutDeepSeekRequestExample1.yaml

@@ -0,0 +1,80 @@
+summary: A completion task
+description: Run `PUT _inference/completion/deepseek-completion` to create an inference endpoint that performs a completion task.
+method_request: 'PUT _inference/completion/deepseek-completion'
+# type: "request"
+value: |-
+  {
+      "service": "deepseek",
+      "service_settings": {
+          "api_key": "DeepSeek-Api-key",
+          "model_id": "deepseek-chat"
+      }
+  }
+alternatives:
+  - language: Python
+    code: |-
+      resp = client.inference.put(
+          task_type="completion",
+          inference_id="deepseek-completion",
+          inference_config={
+              "service": "deepseek",
+              "service_settings": {
+                  "api_key": "DeepSeek-Api-key",
+                  "model_id": "deepseek-chat"
+              }
+          },
+      )
+  - language: JavaScript
+    code: |-
+      const response = await client.inference.put({
+        task_type: "completion",
+        inference_id: "deepseek-completion",
+        inference_config: {
+          service: "deepseek",
+          service_settings: {
+            api_key: "DeepSeek-Api-key",
+            model_id: "deepseek-chat"
+          },
+        },
+      });
+  - language: Ruby
+    code: |-
+      response = client.inference.put(
+        task_type: "completion",
+        inference_id: "deepseek-completion",
+        body: {
+          "service": "deepseek",
+          "service_settings": {
+            "api_key": "DeepSeek-Api-key",
+            "model_id": "deepseek-chat"
+          }
+        }
+      )
+  - language: PHP
+    code: |-
+      $resp = $client->inference()->put([
+          "task_type" => "completion",
+          "inference_id" => "deepseek-completion",
+          "body" => [
+              "service" => "deepseek",
+              "service_settings" => [
+                  "api_key" => "DeepSeek-Api-key",
+                  "model_id" => "deepseek-chat"
+              ],
+          ],
+      ]);
+  - language: curl
+    code:
+      'curl -X PUT -H "Authorization: ApiKey $ELASTIC_API_KEY" -H "Content-Type: application/json" -d
+      ''{"service":"deepseek","service_settings":{"api_key":"DeepSeek-Api-key","model_id":"deepseek-chat"
+      }'' "$ELASTICSEARCH_URL/_inference/completion/deepseek-completion"'
+  - language: Java
+    code: >
+      client.inference().put(p -> p
+          .inferenceId("deepseek-completion")
+          .taskType(TaskType.Completion)
+          .inferenceConfig(i -> i
+              .service("deepseek")
+              .serviceSettings(JsonData.fromJson("{\"api_key\":\"DeepSeek-Api-key\",\"model_id\":\"deepseek-chat\"}"))
+          )
+      );