/telnyx-ai-inference-javascript

name: telnyx-ai-inference-javascript
description: >-
  Access Telnyx LLM inference APIs, embeddings, and AI analytics for call
  insights and summaries. This skill provides JavaScript SDK examples.
metadata:
  author: telnyx
  product: ai-inference
  language: javascript
  generated_by: telnyx-openapi-pipeline

<!-- Auto-generated from Telnyx OpenAPI specs. Do not edit. -->

Telnyx Ai Inference - JavaScript

Installation

npm install telnyx@6.74.2

Setup

import Telnyx from 'telnyx';

const client = new Telnyx({
  apiKey: process.env['TELNYX_API_KEY'], // This is the default and can be omitted
});

All examples below assume `client` is already initialized as shown above.

Error Handling

All API calls can fail with network errors, rate limits (429), validation errors (422), or authentication errors (401). Always handle errors in production code:

try {
  const result = await client.messages.send({ to: '+13125550001', from: '+13125550002', text: 'Hello' });
} catch (err) {
  if (err instanceof Telnyx.APIConnectionError) {
    console.error('Network error — check connectivity and retry');
  } else if (err instanceof Telnyx.RateLimitError) {
    // 429: rate limited — wait and retry with exponential backoff
    const retryAfter = err.headers?.['retry-after'] || 1;
    await new Promise(r => setTimeout(r, retryAfter * 1000));
  } else if (err instanceof Telnyx.APIError) {
    console.error(`API error ${err.status}: ${err.message}`);
    if (err.status === 422) {
      console.error('Validation error — check required fields and formats');
    }
  }
}

Common error codes: `401` invalid API key, `403` insufficient permissions, `404` resource not found, `422` validation error (check field formats), `429` rate limited (retry with exponential backoff).

Important Notes

• **Pagination:** List methods return an auto-paginating iterator. Use `for await (const item of result) { ... }` to iterate through all pages automatically.

Transcribe speech to text

Transcribe speech to text. This endpoint is consistent with the [OpenAI Transcription API](https://platform.openai.com/docs/api-reference/audio/createTranscription) and may be used with the OpenAI JS or Python SDK.

`POST /ai/audio/transcriptions`

import fs from 'fs';

const response = await client.ai.audio.transcribe({ model: 'distil-whisper/distil-large-v2' });

console.log(response.text);

Returns: `duration` (number), `segments` (array[object]), `text` (string), `words` (array[object])

Create a chat completion

**Deprecated**: Use `POST /v2/ai/openai/chat/completions` instead. Chat with a language model. This endpoint is consistent with the [OpenAI Chat Completions API](https://platform.openai.com/docs/api-reference/chat) and may be used with the OpenAI JS or Python SDK.

`POST /ai/chat/completions` — Required: `messages`

Optional: `api_key_ref` (string), `best_of` (integer), `early_stopping` (boolean), `enable_thinking` (boolean), `frequency_penalty` (number), `guided_choice` (array[string]), `guided_json` (object), `guided_regex` (string), `length_penalty` (number), `logprobs` (boolean), `max_tokens` (integer), `min_p` (number), `model` (string), `n` (number), `presence_penalty` (number), `response_format` (object), `seed` (integer), `stop` (object), `stream` (boolean), `temperature` (number), `tool_choice` (enum: none, auto, required), `tools` (array[object]), `top_logprobs` (integer), `top_p` (number), `use_beam_search` (boolean)

const response = await client.ai.chat.createCompletion({
  messages: [
    { role: 'system', content: 'You are a friendly chatbot.' },
    { role: 'user', content: 'Hello, world!' },
  ],
});

console.log(response);

List conversations

Retrieve a list of all AI conversations configured by the user. Supports [PostgREST-style query parameters](https://postgrest.org/en/stable/api.html#horizontal-filtering-rows) for filtering. Examples are included for the standard metadata fields, but you can filter on any field in the metadata JSON object.

`GET /ai/conversations`

const conversations = await client.ai.conversations.list();

console.log(conversations.data);

Returns: `created_at` (date-time), `id` (uuid), `last_message_at` (date-time), `metadata` (object), `name` (string)

Create a conversation

Create a new AI Conversation.

`POST /ai/conversations`

Optional: `metadata` (object), `name` (string)

const conversation = await client.ai.conversations.create();

console.log(conversation.id);

Returns: `created_at` (date-time), `id` (uuid), `last_message_at` (date-time), `metadata` (object), `name` (string)

Aggregate Conversation Insights

Aggregate conversation insights by specified fields

`GET /ai/conversations/conversation-insights/aggregates`

const response = await client.ai.conversations.conversationInsights.aggregate();

console.log(response.data);

Returns: `record_count` (integer)

Get Insight Template Groups

Get all insight groups

`GET /ai/conversations/insight-groups`

// Automatically fetches more pages as needed.
for await (const insightTemplateGroup of client.ai.conversations.insightGroups.retrieveInsightGroups()) {
  console.log(insightTemplateGroup.id);
}

Returns: `created_at` (date-time), `description` (string), `id` (uuid), `insights` (array[object]), `name` (string), `webhook` (string)

Create Insight Template Group

Create a new insight group

`POST /ai/conversations/insight-groups` — Required: `name`

Optional: `description` (string), `webhook` (string)

const insightTemplateGroupDetail = await client.ai.conversations.insightGroups.insightGroups({
  name: 'my-resource',
});

console.log(insightTemplateGroupDetail.data);

Returns: `created_at` (date-time), `description` (string), `id` (uuid), `insights` (array[object]), `name` (string), `webhook` (string)

Get Insight Template