Phase 9: Public Access Live

One Key. All Models.
Zero Compromise.

Access GPT-4 compatible endpoints with automatic failover to Hugging Face and Cohere. Reliable, open-source, and privacy-focused.

Start Building Read Documentation

Unified API

One simple interface for all top models. No more juggling different SDKs or API signatures.

Instant Streaming

Built for speed with native streaming support. We optimize routing for the lowest latency.

No Model Hunting

Stop chasing the latest model weights. We auto-integrate the simplified best-in-class models.

Integration in Seconds

Switching to Freeference is as easy as changing a single line of code.

Get your Key

Update Base URL

Point your OpenAI SDK to https://api.freeference.com/v1.

Deploy

Run your app. We handle the model routing and failover.

example.ts

<span className="text-pink-400">import</span> OpenAI <span className="text-pink-400">from</span> <span className="text-green-300">'openai'</span>;

<span className="text-gray-500">// Initialize with Freeference</span>
<span className="text-pink-400">const</span> client = <span className="text-pink-400">new</span> OpenAI({
  apiKey: <span className="text-green-300">process.env.FREEFERENCE_KEY</span>,
  baseURL: <span className="text-green-300">'https://api.freeference.io/v1'</span>
});

<span className="text-pink-400">const</span> response = <span className="text-pink-400">await</span> client.chat.completions.create({
  model: <span className="text-green-300">'gpt-4-turbo'</span>, <span className="text-gray-500">// We route this!</span>
  messages: [{ role: <span className="text-green-300">'user'</span>, content: <span className="text-green-300">'Hello world!'</span> }]
});

Ready to break free?

Join thousands of developers building the next generation of AI apps without vendor lock-in.

Get Your API Key

One Key. All Models. Zero Compromise.