Skip to contentATP Token

Overview

On this page

ATP is a unified API that gives you access to many AI models through a single endpoint, with provider fallbacks and billing in one place.

Quickstart

Six steps from adding credits to finding your first request in the Console.

API reference

Endpoints, parameters, responses, and error codes.

SDKs and coding agents

Point the OpenAI, Anthropic, or Google SDK — or Claude Code and Codex — at the Gateway.

Use the Console

Set up organizations, projects, and keys, then track usage and billing.

Send your first request#

Replace MODEL_ID with a model id from GET /v1/models, or pick one from the model menu above the example. You need a project API key in $ATP_API_KEY first; the Quickstart walks through it.

curl
Try in Studio
curl https://api.atptoken.ai/v1/chat/completions \
  -H "Authorization: Bearer $ATP_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "MODEL_ID",
    "messages": [{"role": "user", "content": "hi"}]
  }'

What you get#

  • One endpoint, many models. Discover them with GET /v1/models and call any your project allows.
  • Three wire formats. Use the OpenAI, Anthropic, or Google GenAI shape unchanged — only the base URL changes.
  • Automatic fallback. Each model is served by a provider pool, so a single model id keeps working when one provider is degraded. See Provider routing.
  • One bill. Usage across every model and provider is metered in credits. See How credits work.

Three ways to integrate#

ApproachBest for
APIFull control, any language, no dependencies
SDKsType-safe calls with your existing OpenAI / Anthropic / Google SDK
Coding agentsClaude Code, Codex, and other agents that speak the wire format

Articles on AI cost and gateways#

Next steps

NextQuickstart
Was this page helpful?