OpenAI Adapter

The OpenAI adapter wraps the official openai npm package. It's the most feature-complete adapter, supporting all four modalities: text, images, audio, and video analysis.

Installation

Code

npm install tekimax-ts

Usage

Code

import { Tekimax, OpenAIProvider } from 'tekimax-ts';

const client = new Tekimax({
  provider: new OpenAIProvider({
    apiKey: process.env.OPENAI_API_KEY!,
    // Enable browser-side usage (see note below).
    // dangerouslyAllowBrowser: true
  })
});

const result = await client.text.chat.completions.create({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'Hello OpenAI' }]
});

console.log(result.message.content);

Streaming

Code

const stream = client.text.chat.completions.createStream({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'Tell me a story' }]
});

for await (const chunk of stream) {
  process.stdout.write(chunk.delta);
}

Image Generation

Code

const result = await client.images.generate({
  prompt: 'A futuristic cityscape at sunset',
  model: 'dall-e-3',
  size: '1024x1024',
  quality: 'hd'
});

console.log(result.data[0].url);

Image Analysis (Vision)

Code

const analysis = await client.images.analyze({
  model: 'gpt-4o',
  image: 'https://example.com/chart.png',
  prompt: 'Extract the key data points from this chart.'
});

console.log(analysis.content);

Text-to-Speech

Code

const audio = await client.audio.speak({
  model: 'tts-1',
  input: 'Hello, this is a test.',
  voice: 'alloy'
});

console.log(`Generated ${audio.buffer.byteLength} bytes of audio`);

Audio Transcription

Code

const transcription = await client.audio.transcribe({
  file: audioBlob, // File, Blob, or Buffer
  model: 'whisper-1',
  response_format: 'verbose_json'
});

console.log(transcription.text);
console.log(transcription.segments); // Word-level timestamps

Notes

dangerouslyAllowBrowser: The OpenAI SDK blocks client-side usage by default to prevent API key exposure in browser bundles. Set this to true only for prototyping — in production, proxy requests through your backend.
max_completion_tokens: The adapter maps Tekimax's maxTokens to OpenAI's max_completion_tokens field (not the deprecated max_tokens). This is required by OpenAI's v6 SDK.
Stream Options: The adapter automatically sets stream_options: { include_usage: true } on streaming requests so that token usage data is included in the final chunk.

OpenAI Adapter

Installation

Usage

Streaming

Image Generation

Image Analysis (Vision)

Text-to-Speech

Audio Transcription

Notes

On this page