OpenRouter Proxy

This Apify Actor creates a proxy for the Open Router API, allowing you to access multiple AI models through a unified OpenAI-compatible interface. All requests are charged to your Apify account on a pay-per-event basis.

What this Actor does

• Proxy access: Routes your API requests to Open Router's extensive collection of AI models
• OpenAI compatibility: Works seamlessly with the OpenAI SDK and any OpenAI-compatible client
• Transparent billing: Charges are applied to your Apify account at the same rates as Open Router
• Full feature support: Supports both streaming and non-streaming responses
• No API key management: Uses your Apify token for authentication - no need to manage separate Open Router API keys
• Standby mode: Actor runs in standby mode only and has a static URL address, like a standard web server.

Pricing

This Actor uses a pay-per-event pricing model through Apify. Each API request counts as one event, and no additional fees are applied beyond the standard Apify event pricing.

Quick start

1. Install the OpenAI package

$npm install openai

2. Basic usage

import OpenAI from 'openai';

const openai = new OpenAI({
  baseURL: 'https://openrouter.apify.actor/api/v1',
  apiKey: 'no-key-required-but-must-not-be-empty', // Any non-empty string works; do NOT use a real API key.
  defaultHeaders: {
    Authorization: `Bearer ${process.env.APIFY_TOKEN}`, // Apify token is loaded automatically in runtime
  },
});

async function main() {
  const completion = await openai.chat.completions.create({
    model: 'openrouter/auto',
    messages: [
      {
        role: 'user',
        content: 'What is the meaning of life?',
      },
    ],
  });

  console.log(completion.choices[0].message);
}

await main();

3. Streaming responses

const stream = await openai.chat.completions.create({
  model: 'openrouter/auto',
  messages: [
    {
      role: 'user',
      content: 'Write a short story about a robot.',
    },
  ],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || '');
}

4. Image generation with vision models

OpenRouter supports image generation through models with vision and image generation capabilities. Use the chat completions API with the modalities parameter:

const response = await openai.chat.completions.create({
  model: 'google/gemini-2.5-flash-image', // Vision-capable model
  messages: [
    {
      role: 'user',
      content: 'Generate an image of a cute baby sea otter',
    },
  ],
  modalities: ['text', 'image'], // Enable image generation
});

// Access generated image from response
console.log(response.choices[0].message);

Note: OpenRouter doesn't support the traditional /images/generations endpoint. Image generation is done through compatible models using the chat completions API. Check available models with image capabilities on OpenRouter's models page.

Available models

This proxy supports all models available through Open Router from providers, including:

• OpenAI
• Anthropic
• Google
• Meta
• Perplexity
• And many more...

For a complete list of available models, visit Open Router's models page.

Authentication

The Actor uses your Apify token for authentication. In Apify Actor environments, APIFY_TOKEN is automatically available. For local development, you can:

1. Set the environment variable: export APIFY_TOKEN=your_token_here
2. Or pass it directly in the Authorization header
3. Find your token in the Apify Console

Support

For issues related to this Actor, please contact the Actor developer.