Telnyx | AI SDK

Telnyx provides an AI SDK provider for the Vercel AI SDK, giving developers access to chat models, embeddings, speech-to-text, and text-to-speech through a single package.

The Telnyx provider is a good fit if you want:

One package for multiple AI capabilities: chat, embeddings, transcription, and speech generation
OpenAI-compatible model access: use Telnyx-hosted models through a familiar API pattern
Speech support in the same provider: combine text and voice workflows in one integration
Simple setup: standard API key configuration with the AI SDK

Learn more in the Telnyx documentation.

Setup

The @telnyx/ai-sdk-provider package (v2.0.0+) supports AI SDK 7 and the Provider V4 specification.

Install it alongside AI SDK 7:

$ pnpm add @telnyx/ai-sdk-provider ai@^7 zod

$ npm install @telnyx/ai-sdk-provider ai@^7 zod

$ yarn add @telnyx/ai-sdk-provider ai@^7 zod

$ bun add @telnyx/ai-sdk-provider ai@^7 zod

Set your Telnyx API key as an environment variable:

export TELNYX_API_KEY="your_api_key_here"

Provider Instance

To create a Telnyx provider instance, use the createTelnyx function:

import { createTelnyx } from '@telnyx/ai-sdk-provider';

const telnyx = createTelnyx({
  apiKey: process.env.TELNYX_API_KEY,
});

You can also use the default telnyx export when TELNYX_API_KEY is set in your environment.

Examples

`generateText`

import { telnyx } from '@telnyx/ai-sdk-provider';
import { generateText } from 'ai';

const { text } = await generateText({
  model: telnyx('Qwen/Qwen3-235B-A22B'),
  prompt: 'Explain WebRTC in simple terms.',
});

console.log(text);

`embed`

import { telnyx } from '@telnyx/ai-sdk-provider';
import { embed } from 'ai';

const { embedding } = await embed({
  model: telnyx.embeddingModel('thenlper/gte-large'),
  value: 'What is WebRTC?',
});

console.log(embedding);

`transcribe`

import { telnyx } from '@telnyx/ai-sdk-provider';
import { transcribe } from 'ai';
import { readFile } from 'node:fs/promises';

const result = await transcribe({
  model: telnyx.transcriptionModel('distil-whisper/distil-large-v2'),
  audio: await readFile('./audio.wav'),
});

console.log(result.text);

`generateSpeech`

import { telnyx } from '@telnyx/ai-sdk-provider';
import { generateSpeech } from 'ai';

const { audio } = await generateSpeech({
  model: telnyx.speech('Telnyx.NaturalHD.astra'),
  text: 'Hello from Telnyx!',
  voice: 'Telnyx.NaturalHD.astra',
});

await audio.writeToFile('./speech.mp3');