Interactive AI Model Explorer
Want to calculate your exact parameters and operational expenses? Run the calculations locally inside your browser.
Launch AI Model Explorer1. Endpoint Configuration Directory (2026)
| Provider | Base URL | Top Free Model ID | Key Header |
|---|---|---|---|
| **Groq Cloud** | `https://api.groq.com/openai/v1` | `llama-3.3-70b-versatile` | `Authorization: Bearer gsk_...` |
| **OpenRouter** | `https://openrouter.ai/api/v1` | `meta-llama/llama-3.3-70b-instruct:free` | `Authorization: Bearer sk-or-...` |
| **GitHub Models** | `https://models.inference.ai.azure.com` | `gpt-4o-mini` | `Authorization: Bearer ghp_...` |
| **Mistral AI** | `https://api.mistral.ai/v1` | `mistral-small-latest` | `Authorization: Bearer ...` |
| **Together AI** | `https://api.together.xyz/v1` | `meta-llama/Llama-3.3-70B-Instruct-Turbo` | `Authorization: Bearer ...` |
2. Multi-Provider Client Factory Pattern (TypeScript)
Create a factory function to route traffic across providers dynamically:
```typescript import OpenAI from 'openai';
export type ProviderName = 'groq' | 'openrouter' | 'github';
export function getLLMClient(provider: ProviderName = 'groq') { const configs = { groq: { baseURL: 'https://api.groq.com/openai/v1', apiKey: process.env.GROQ_API_KEY, defaultModel: 'llama-3.3-70b-versatile', }, openrouter: { baseURL: 'https://openrouter.ai/api/v1', apiKey: process.env.OPENROUTER_API_KEY, defaultModel: 'meta-llama/llama-3.3-70b-instruct:free', }, github: { baseURL: 'https://models.inference.ai.azure.com', apiKey: process.env.GITHUB_TOKEN, defaultModel: 'gpt-4o-mini', }, };
const config = configs[provider]; return { client: new OpenAI({ baseURL: config.baseURL, apiKey: config.apiKey }), model: config.defaultModel, }; } ```
Frequently Asked Questions
Do OpenAI-compatible endpoints support streaming?
Yes. All standard OpenAI clients (stream: true) work out of the box with Server-Sent Events (SSE) across Groq, OpenRouter, and GitHub Models.
Can I use curl to test these endpoints?
Yes. Simply send a standard POST request with a JSON body to the provider’s /chat/completions path.