What is PlainField

PlainField is a hosted LLM gateway and observability platform. You point your application at one endpoint, and PlainField proxies the request to any provider (OpenAI, Anthropic, Google, and more), logs it, estimates its cost, and shows that data in your dashboard.

You do not run or operate anything. Create an account, generate an API key, and send requests — PlainField runs the gateway for you at https://api.plainfield-ai.com.

The fastest path

Send your first request in under five minutes:


curl -X POST https://api.plainfield-ai.com/ai/chat/completions \
  -H 'Authorization: Bearer pt_live_...' \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "openai/gpt-4o-mini",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Replace pt_live_... with a key you create in the dashboard. Full walkthrough: Quickstart.

How it works

Your app sends an OpenAI-format request to the gateway with your PlainField API key.
The gateway forwards it to the right provider and returns the response.
In the background, PlainField records the request — tokens, latency, and estimated cost — and makes it visible in your dashboard.

What you get

Capability	What it means for you
One endpoint	Call 100+ models from many providers through a single OpenAI-compatible API.
Cost tracking	Every request is priced automatically. See spend by model and prompt in your dashboard.
Observability	Latency, token counts, and full request logs in your dashboard.
Your keys or ours	Use PlainField’s managed provider access, or bring your own provider keys.

Find your way

New here? Start with the Quickstart.
Learning by doing? Follow a Tutorial.
Have a specific goal? Jump to a How-to guide.
Need exact details? See the Reference.
Want the why? Read the Explanation.

Attribution

The gateway is built on the Helicone AI Gateway (GPL v3.0), with cost estimation, API-key authentication, and a logs and costs REST API added.