OpenAI SDK Compatibility

Vortik Gateway acts as a drop-in stand-in for any app using the OpenAI SDK. Simply override baseURL and apiKey, and you instantly unlock Vortik's capabilities without changing your core integration logic.

Multi-Provider Support

One of Vortik's biggest strengths is routing to different LLM providers through that single, familiar OpenAI interface. Whether you want to call an OpenAI, Azure, Anthropic, or Google model, you can route all of them seamlessly.

Routing to Different Providers

Before routing, make sure your provider keys are configured in the LLM Keys section of the Vortik Dashboard.

Once that's set, just use the model name when making your request (e.g., prefixing it with the provider like anthropic/):

Node.js

const response = await openai.chat.completions.create({
  model: 'anthropic/claude-3-opus', // Routing to Anthropic through Vortik
  messages: [{ role: 'user', content: 'Tell me about the universe.' }],
});

Python

response = client.chat.completions.create(
    model="anthropic/claude-3-opus",
    messages=[{"role": "user", "content": "Tell me about the universe."}]
)

.NET

ChatCompletion completion = client.CompleteChat(
    [new UserChatMessage("Tell me about the universe.")],
    new ChatCompletionOptions { Model = "anthropic/claude-3-opus" }
);

PHP

$response = $client->chat()->create([
    'model' => 'anthropic/claude-3-opus',
    'messages' => [
        ['role' => 'user', 'content' => 'Tell me about the universe.'],
    ],
]);

cURL

curl "https://eu.gateway.vortik.ai/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_VORTIK_API_KEY" \
  -d '{
    "model": "anthropic/claude-3-opus",
    "messages": [{"role": "user", "content": "Tell me about the universe."}]
  }'

---

Next, learn how to keep your application logic clean with Prompt Injection.