inference refactor (#4153)

Author: l-trotta

output/openapi/elasticsearch-openapi.json

@@ -18,18 +18,17 @@
  */
 
 import { RequestChatCompletionBase } from '@inference/_types/CommonTypes'
-import { UserDefinedValue } from '@spec_utils/UserDefinedValue'
+import { RequestBase } from '@_types/Base'
 import { Id } from '@_types/common'
 import { Duration } from '@_types/Time'
-
 /**
  * Perform chat completion inference
  * @rest_spec_name inference.chat_completion_unified
  * @availability stack since=8.18.0 stability=stable visibility=public
  * @availability serverless stability=stable visibility=public
  * @doc_id inference-api-chat-completion
  */
-export interface Request extends RequestChatCompletionBase {
+export interface Request extends RequestBase {
   urls: [
     {
       path: '/_inference/chat_completion/{inference_id}/_stream'
@@ -49,144 +48,6 @@ export interface Request extends RequestChatCompletionBase {
      */
     timeout?: Duration
   }
-}
-
-/**
- * @codegen_names string, object
- */
-export type CompletionToolType = string | CompletionToolChoice
-
-/**
- * An object style representation of a single portion of a conversation.
- */
-export interface ContentObject {
-  /**
-   * The text content.
-   */
-  text: string
-  /**
-   * The type of content.
-   */
-  type: string
-}
-
-/**
- * The function that the model called.
- */
-export interface ToolCallFunction {
-  /**
-   * The arguments to call the function with in JSON format.
-   */
-  arguments: string
-  /**
-   * The name of the function to call.
-   */
-  name: string
-}
-
-/**
- * A tool call generated by the model.
- */
-export interface ToolCall {
-  /**
-   * The identifier of the tool call.
-   */
-  id: Id
-  /**
-   * The function that the model called.
-   */
-  function: ToolCallFunction
-  /**
-   * The type of the tool call.
-   */
-  type: string
-}
-
-/**
- * @codegen_names string, object
- */
-export type MessageContent = string | Array<ContentObject>
-
-/**
- * An object representing part of the conversation.
- */
-export interface Message {
-  /**
-   * The content of the message.
-   */
-  content?: MessageContent
-  /**
-   * The role of the message author.
-   */
-  role: string
-  /**
-   * The tool call that this message is responding to.
-   */
-  tool_call_id?: Id
-  /**
-   * The tool calls generated by the model.
-   */
-  tool_calls?: Array<ToolCall>
-}
-
-/**
- * The tool choice function.
- *
- */
-export interface CompletionToolChoiceFunction {
-  /**
-   * The name of the function to call.
-   */
-  name: string
-}
-
-/**
- * Controls which tool is called by the model.
- */
-export interface CompletionToolChoice {
-  /**
-   * The type of the tool.
-   */
-  type: string
-  /**
-   * The tool choice function.
-   */
-  function: CompletionToolChoiceFunction
-}
-
-/**
- * The completion tool function definition.
- */
-export interface CompletionToolFunction {
-  /**
-   * A description of what the function does.
-   * This is used by the model to choose when and how to call the function.
-   */
-  description?: string
-  /**
-   * The name of the function.
-   */
-  name: string
-  /**
-   * The parameters the functional accepts. This should be formatted as a JSON object.
-   */
-  parameters?: UserDefinedValue
-  /**
-   * Whether to enable schema adherence when generating the function call.
-   */
-  strict?: boolean
-}
-
-/**
- * A list of tools that the model can call.
- */
-export interface CompletionTool {
-  /**
-   * The type of tool.
-   */
-  type: string
-  /**
-   * The function definition.
-   */
-  function: CompletionToolFunction
+  /** @codegen_name chat_completion_request */
+  body: RequestChatCompletionBase
 }

output/openapi/elasticsearch-serverless-openapi.json

@@ -18,8 +18,8 @@
  */
 
 import { RequestChatCompletionBase } from '@inference/_types/CommonTypes'
+import { RequestBase } from '@_types/Base'
 import { Id } from '@_types/common'
-
 /**
  * Perform a chat completion task through the Elastic Inference Service (EIS).
  *
@@ -30,7 +30,7 @@ import { Id } from '@_types/common'
  * @cluster_privileges manage_inference
  * @doc_id inference-api-post-eis-chat-completion
  */
-export interface Request extends RequestChatCompletionBase {
+export interface Request extends RequestBase {
   urls: [
     {
       path: '/_inference/chat_completion/{eis_inference_id}/_stream'
@@ -43,4 +43,6 @@ export interface Request extends RequestChatCompletionBase {
      */
     eis_inference_id: Id
   }
+  /** @codegen_name chat_completion_request */
+  body: RequestChatCompletionBase
 }

output/schema/schema-serverless.json

@@ -18,9 +18,12 @@
  */
 
 import {
-  InferenceChunkingSettings,
-  RateLimitSetting
-} from '@inference/_types/Services'
+  AlibabaCloudServiceSettings,
+  AlibabaCloudServiceType,
+  AlibabaCloudTaskSettings,
+  AlibabaCloudTaskType
+} from '@inference/_types/CommonTypes'
+import { InferenceChunkingSettings } from '@inference/_types/Services'
 import { RequestBase } from '@_types/Base'
 import { Id } from '@_types/common'
 
@@ -66,7 +69,7 @@ export interface Request extends RequestBase {
     /**
      * The type of service supported for the specified task type. In this case, `alibabacloud-ai-search`.
      */
-    service: ServiceType
+    service: AlibabaCloudServiceType
     /**
      * Settings used to install the inference model. These settings are specific to the `alibabacloud-ai-search` service.
      */
@@ -78,77 +81,3 @@ export interface Request extends RequestBase {
     task_settings?: AlibabaCloudTaskSettings
   }
 }
-
-export enum AlibabaCloudTaskType {
-  completion,
-  rerank,
-  space_embedding,
-  text_embedding
-}
-
-export enum ServiceType {
-  'alibabacloud-ai-search'
-}
-
-export class AlibabaCloudServiceSettings {
-  /**
-   * A valid API key for the AlibabaCloud AI Search API.
-   */
-  api_key: string
-  /**
-   * The name of the host address used for the inference task.
-   * You can find the host address in the API keys section of the documentation.
-   * @ext_doc_id alibabacloud-api-keys
-   */
-  host: string
-  /**
-   * This setting helps to minimize the number of rate limit errors returned from AlibabaCloud AI Search.
-   * By default, the `alibabacloud-ai-search` service sets the number of requests allowed per minute to `1000`.
-   */
-  rate_limit?: RateLimitSetting
-  /**
-   * The name of the model service to use for the inference task.
-   * The following service IDs are available for the `completion` task:
-   *
-   * * `ops-qwen-turbo`
-   * * `qwen-turbo`
-   * * `qwen-plus`
-   * * `qwen-max ÷ qwen-max-longcontext`
-   *
-   * The following service ID is available for the `rerank` task:
-   *
-   * * `ops-bge-reranker-larger`
-   *
-   * The following service ID is available for the `sparse_embedding` task:
-   *
-   * * `ops-text-sparse-embedding-001`
-   *
-   * The following service IDs are available for the `text_embedding` task:
-   *
-   * `ops-text-embedding-001`
-   * `ops-text-embedding-zh-001`
-   * `ops-text-embedding-en-001`
-   * `ops-text-embedding-002`
-   */
-  service_id: string
-  /**
-   * The name of the workspace used for the inference task.
-   */
-  workspace: string
-}
-
-export class AlibabaCloudTaskSettings {
-  /**
-   * For a `sparse_embedding` or `text_embedding` task, specify the type of input passed to the model.
-   * Valid values are:
-   *
-   * * `ingest` for storing document embeddings in a vector database.
-   * * `search` for storing embeddings of search queries run against a vector database to find relevant documents.
-   */
-  input_type?: string
-  /**
-   * For a `sparse_embedding` task, it affects whether the token name will be returned in the response.
-   * It defaults to `false`, which means only the token ID will be returned in the response.
-   */
-  return_token?: boolean
-}