Hypereal TypeScript SDK

Official TypeScript SDK for the Hypereal AI API. Use one client for LLM responses, OpenAI-compatible chat completions, image generation, video generation, audio generation, job polling, and storage uploads.

Install

pnpm add hypereal-sdk

Usage

import { Hypereal } from 'hypereal-sdk';

const client = new Hypereal({
  apiKey: process.env.HYPEREAL_API_KEY!,
});

const models = await client.models.list();
console.log(models.data.map((model) => model.id));

Chat completions

const completion = await client.chat.completions.create({
  model: 'gpt-5.5',
  messages: [
    { role: 'system', content: 'You write concise production notes.' },
    { role: 'user', content: 'Summarize the image generation API.' },
  ],
});

Responses API

const response = await client.responses.create({
  model: 'gpt-5.5',
  input: 'Write a short launch checklist for an SDK release.',
});

Image generation

const image = await client.images.generate({
  model: 'gemini-3-1-flash-t2i',
  prompt: 'A cinematic portrait in neon light',
  aspect_ratio: '16:9',
  resolution: '2K',
});

Video generation

const video = await client.videos.generate({
  model: 'veo-3.1',
  input: {
    prompt: 'A product shot rotating on a clean studio table',
    duration: 5,
  },
});

Audio generation

const audio = await client.audio.generate({
  model: 'audio-tts',
  input: {
    text: 'Ship the SDK with typed examples and clear errors.',
    voice: 'alloy',
  },
});

Poll jobs

Some media providers return a job id. Poll until a terminal response is ready:

const job = await client.jobs.poll('job_or_prediction_id', {
  params: {
    model: 'veo-3.1',
    type: 'video',
  },
  intervalMs: 2500,
  timeoutMs: 180_000,
});

Storage uploads

const result = await client.storage.uploadFile(file, {
  filename: 'training-image.png',
  contentType: 'image/png',
  kind: 'dataset',
});

const object = await client.storage.get(result.id);

For local development, point the SDK at your Next.js server:

const client = new Hypereal({
  apiKey: process.env.HYPEREAL_API_KEY!,
  baseURL: 'http://localhost:3000/api',
});

Streaming

Streaming endpoints return the raw Response so callers can consume the body:

const response = await client.chat.completions.create<Response>({
  model: 'gpt-5.5',
  stream: true,
  messages: [{ role: 'user', content: 'Stream a short answer.' }],
});

for await (const chunk of response.body!) {
  process.stdout.write(Buffer.from(chunk).toString('utf8'));
}

API reference

• client.models.list()
• client.chat.completions.create(body)
• client.responses.create(body)
• client.images.generate(body)
• client.videos.generate(body)
• client.audio.generate(body)
• client.jobs.get(id, params)
• client.jobs.poll(id, options)
• client.storage.createUploadUrl(body)
• client.storage.uploadFile(file, body)
• client.storage.commit(id)
• client.storage.list(params)
• client.storage.get(id)
• client.storage.update(id, { isPublic })
• client.storage.delete(id)

Errors

import { Hypereal, HyperealAPIError } from 'hypereal-sdk';

try {
  await new Hypereal({ apiKey: 'ck_invalid' }).models.list();
} catch (error) {
  if (error instanceof HyperealAPIError) {
    console.log(error.status, error.type, error.message);
  }
}