AI SDK - AIMLAPI Provider

The AIMLAPI provider for the AI SDK enables access to over 300+ models via the AIMLAPI platform.

Installation

Install the official package:

npm i vercel-ai-aimlapi
# or
yarn add vercel-ai-aimlapi
# or
pnpm add vercel-ai-aimlapi

Usage

Import the default provider instance:

import { aimlapi } from 'vercel-ai-aimlapi';

Text Generation Example

import { aimlapi } from 'vercel-ai-aimlapi';
import { generateText } from 'ai';

const { text } = await generateText({
  model: aimlapi('gpt-3.5-turbo'),
  prompt: 'Hello from AIMLAPI!',
});

console.log(text);

Advanced Usage: gpt-4o

import 'dotenv/config';
import { generateText, LanguageModel } from 'ai';
import { aimlapi } from 'vercel-ai-aimlapi';

async function main() {
  const { text } = await generateText({
    model: aimlapi('gpt-4o') as LanguageModel,
    system: 'You are a friendly assistant!',
    prompt: 'Why is the sky blue?',
  });

  console.log(text);
}

main();

Image Generation Example

import 'dotenv/config';
import { experimental_generateImage } from 'ai';
import { aimlapi } from 'vercel-ai-aimlapi';
import * as fs from 'fs';

async function main() {
  const model = aimlapi.imageModel?.('flux-realism');
  if (!model) throw new Error('Image model not available');

  const { image } = await experimental_generateImage({
    model,
    prompt: 'pixelated image of a cute baby sea otter, 8-bit style, vibrant colors',
  });

  const jsonText = new TextDecoder().decode(image.uint8Array);
  const json = JSON.parse(jsonText);
  const url = json.data?.[0]?.url;

  if (!url) {
    console.error('❌ URL not found in image response');
    return;
  }

  console.log('✅ Image URL:', url);

  fs.writeFile('imageUrlFile.txt', url, (err) => {
    if (err) {
      console.error('Error writing to file:', err);
      return;
    }
    console.log('File written successfully.');
  });
}

main().catch(console.error);

Documentation

• 📘 AIMLAPI Provider Docs
• 🔗 AIMLAPI Platform