@orquesta/js-sdk v1.3.3

LLM Operations and Integration Platform

Orquesta Javascript SDK

Contents

• Installation
• Create a client instance
• Usage
• Reference

Installation

npm install @orquesta/js-sdk

yarn add @orquesta/js-sdk

Creating a client instance

You can get your workspace API key from the settings section in your Orquesta workspace. https://my.orquesta.dev/<workspace>/settings/developers

Initialize the Orquesta client with your API key:

import { createClient } from '@orquesta/node';

const client = createClient({
  api_key: '__API_KEY__',
  ttl: 3600,
});

When creating a client instance, the following connection settings can be adjusted:

• api_key: string - your workspace API key to use for authentication.
• environment?: string - the environment to use for the client. Not required but recommended to use so it's added to the evaluation context automatically.
• ttl?: number - the time to live in seconds for the local cache. Default is 3600 seconds (1 hour).

Usage - Endpoints

Use the Endpoints API to query or stream your endpoints from Orquesta.

Using endpoints to generate a LLM response based on your use case with Orquesta provides a low-latency, secure connection to the Endpoints API online prediction service. Getting out of the box metrics and logging for your LLMs.

Endpoints API support streaming and querying. The endpoint response type is OrquestaEndpointResponse. We recommend to use the code snippets provided in the Orquesta Admin panel to reduce risk of errors and improve ease of use.

Example: Querying an endpoint

const completion = await client.endpoints.query({
  key: 'customer_service',
  context: { environments: 'production', country: 'NLD' },
  variables: { firstname: 'John', city: 'New York' },
  metadata: { customer_id: 'Qwtqwty90281' },
});

console.log(completion.content);

Example: Streaming your endpoints

We provide support for streaming responses using Server Sent Events (SSE). The stream method output a RxJS observable that you can subscribe and receive the streaming chunks of data from the endpoint. The chunk type is OrquestaEndpointResponse

let endpointRef: OrquestaEndpoint;

const stream = client.endpoints
  .stream({
    key: 'customer_service',
    context: { environments: 'production', country: 'NLD' },
    variables: { firstname: 'John', city: 'New York' },
    metadata: { customer_id: 'Qwtqwty90281' },
  })
  .subscribe({
    next: (endpoint) => {
      endpointRef = endpoint;
      console.log(endpoint.content);
    },
    complete: () => {
      // Executed when the streaming is completed
    },
    error: (error) => {
      console.log(error);
    },
  });

If you need to cancel a stream, you can call stream.unsubscribe() method.

Logging score and metadata for endpoints

After every query, Orquesta will generate a log with the result of the evaluation. You can add metadata and score to the endpoint by using the addMetrics method.

endpointRef.addMetrics({
  score: 85,
  metadata: {
    custom: 'custom_metadata',
    chain_id: 'ad1231xsdaABw',
  },
});

Usage - Prompts

Use the Prompts API to query your prompts from Orquesta.

You can use Orquesta in prompt management mode by consuming our Prompts API. The prompt value type is OrquestaPrompt. We recommend to use the code snippets provided in the Orquesta Admin panel to reduce risk of errors and improve ease of use.

We support an unified data model structure for all our prompts and provide helper functions that map the returned value from Orquesta to the specific provider.

The query method receives an object of type OrquestaPromptRequest as parameter.

Example: Querying a prompt

import { orquestaOpenAIMapper } from '@orquesta/utils';
import { OrquestaOpenAIPromptParameters, OrquestaPrompt } from '@orquesta/core';

const prompt: OrquestaPrompt = await client.prompts.query({
  key: 'prompt_key',
  context: { environments: 'production', country: 'NLD' },
  variables: { firstname: 'John', city: 'New York' },
  metadata: { chain_id: 'ad1231xsdaABw' },
});

const openAIPrompt: OrquestaOpenAIPromptParameters =
  orquestaOpenAIParametersMapper(prompt.value);

Helper functions per LLM provider

We provide helper functions and interfaces that map the returned value from Orquesta to the specific provider, so it's easy for you to forward the Prompt to your different LLM providers.

Logging responses and metadata for prompts

After every query, Orquesta will generate a log with the result of the evaluation. You can add metadata and information about the interaction with the LLM to the log by using the addMetrics method.

The properties score, latency, llm_response and economics are reserved and used to generate your real-time dashboards. metadata is a set of key-value pairs that you can use to add custom information to the log.

Example: Add metrics to your request log

prompt.addMetrics({
  score: 85,
  latency: 4000,
  llm_response: 'Orquesta is awesome!',
  economics: {
    prompt_tokens: 1200,
    completion_tokens: 750,
    total_tokens: 1950,
  },
  metadata: {
    custom: 'custom_metadata',
    chain_id: 'ad1231xsdaABw',
    total_interactions: 200,
  },
});

Usage - Remote Configurations

Orquesta has a powerful Remote Configurations API that allows you to configure and run all your environments and services remotely dynamically. Orquesta supports different types of remote configurations, and we recommend always typing the query method to help Typescript infer the correct type.

The useOrquestaRemoteConfig hook receives an object of type OrquestaRemoteConfigQuery as parameter.

Supported types: boolean, number, string, json, array

Example: Querying a configuration of type boolean

const config: OrquestaRemoteConfig<boolean> =
  await client.remoteConfigs.query<boolean>({
    key: 'boolean_config',
    default_value: false,
    context: { environments: 'production', role: 'admin' },
    metadata: { timestamp: Date.now() },
  });

Example: Querying a configuration of type string

const config: OrquestaRemoteConfig<string> =
  await client.remoteConfigs.query<string>({
    key: 'string_config',
    default_value: 'string_value',
    context: { environments: 'production', country: 'NL' },
    metadata: { timestamp: Date.now() },
  });

Example: Querying a configuration of type number

const config: OrquestaRemoteConfig<number> =
  await client.remoteConfigs.query<number>({
    key: 'number_config',
    default_value: 1990,
    context: { environments: 'production', market: 'US', domain: 'ecommerce' },
    metadata: { timestamp: Date.now() },
  });

Example: Querying a configuration of type array

const config: OrquestaRemoteConfig<string[]> = await client.remoteConfigs.query<
  string[]
>({
  key: 'array_config',
  default_value: ['value1', 'value2'],
  context: { environments: 'acceptance', isEnable: true },
  metadata: { timestamp: Date.now() },
});

Example: Querying a configuration of type json

It's recommended to add an interface to the json configuration to help Typescript infer the correct type.

const config: OrquestaRemoteConfig<object> =
  await client.remoteConfigs.query<object>({
    key: 'json_config',
    default_value: { dashboardEnabled: false, theme: 'dark' },
    context: { environments: 'develop', platform: 'mobile' },
    metadata: { timestamp: Date.now() },
  });

Additional metadata logging

After every query, Orquesta will generate a log with data about the request. You can add metadata to the log using the addMetrics method anytime.

metadata is a set of key-value pairs that you can use to add custom information to the log.

Example: Add metrics to your request log

prompt.addMetrics({
  metadata: {
    custom: 'custom_metadata',
    user_clicks: 20,
    selected_option: 'option1',
  },
});

Orquesta API

Endpoints API

Types:

• OrquestaEndpoint
• OrquestaEndpointMetrics
• OrquestaEndpointRequest

Methods:

• client.endpoints.query({ ...params }) -> Promise<OrquestaEndpoint>
• client.endpoints.stream({ ...params }) -> Observable<OrquestaEndpoint>

Prompts API

Types:

• OrquestaPrompt
• OrquestaPromptMetrics
• OrquestaPromptMetricsEconomics
• OrquestaPromptRequest

Methods:

• client.prompts.query({ ...params }) -> Promise<OrquestaPrompt>

RemoteConfigs API

Types:

• OrquestaRemoteConfig
• OrquestaRemoteConfigMetrics
• OrquestaRemoteConfigRequest

Methods:

• client.remoteconfigs.query({ ...params }) -> Promise<OrquestaRemoteConfig<T>>