API Overview

Cortecs provides an OpenAI-compatible API that makes it simple to run serverless inference across multiple providers with no infrastructure setup.

The API supports three main capabilities:

🔁 Chat Completions : POST /v1/chat/completions

Send chat requests using any available model. Supports standard OpenAI parameters like messages, temperature, and max_tokens.

Use the preference parameter to optimize routing for speed, cost, or balanced.

🧩 Embeddings : POST /v1/embeddings

Generate text embeddings using supported models. The API is compatible with OpenAI’s embedding request format and can be routed across multiple providers using the same preference-based selection.

📦 Model Listing : GET /v1/models

Retrieve the full list of available models, including their supported features, costs, context sizes, and providers.