Quickstart

When to use this pageCopy Anchor Link

Use this page to go from no account to a working request in a few minutes. It ends with the same request in the Anthropic and OpenAI SDKs.

Before you startCopy Anchor Link

You need a card to add credit, and either curl, Python 3 or Node.js.

Set up your account and keyCopy Anchor Link

export EZAT_API_KEY="ezat_sk_…"

Send your first requestCopy Anchor Link

This calls the Anthropic Messages API with curl:

curl --fail-with-body -sS https://api.router.ezat.io/v1/messages \
  -H "x-api-key: $EZAT_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-5-5",
    "max_tokens": 64,
    "messages": [{"role": "user", "content": "Say hello in five words."}]
  }'

The response is a standard Anthropic message. Its model field is the name you sent, and its usage field has the token counts you are billed for. The request also appears on the Usage page within a few seconds.

If you get an error instead, the body says why. The usual ones are 401 (the key isn't set or isn't right) and 402 (no credit). See Troubleshooting.

List the modelsCopy Anchor Link

curl --fail-with-body -sS https://api.router.ezat.io/v1/models \
  -H "x-api-key: $EZAT_API_KEY"

Each entry's id is what you send as model. Ids have the form author/model, such as anthropic/claude-sonnet-5-5. Anthropic's own model names work too. See Models for every id, its other names and its limits.

Use an SDKCopy Anchor Link

The Anthropic SDKs work unchanged: set the base URL to https://api.router.ezat.io and use your EZAT Router key. Install them with pip install anthropic or npm install @anthropic-ai/sdk.

import os

import anthropic

client = anthropic.Anthropic(
    base_url="https://api.router.ezat.io",
    api_key=os.environ["EZAT_API_KEY"],
)

message = client.messages.create(
    model="anthropic/claude-sonnet-5-5",
    max_tokens=64,
    messages=[{"role": "user", "content": "Say hello in five words."}],
)
print("".join(block.text for block in message.content if block.type == "text"))

With an OpenAI SDKCopy Anchor Link

If your code or tool speaks OpenAI's format, use the OpenAI-compatible base URL, https://api.router.ezat.io/openai/v1, with the same key. Install the SDK with pip install openai or npm install openai.

import os

from openai import OpenAI

client = OpenAI(
    base_url="https://api.router.ezat.io/openai/v1",
    api_key=os.environ["EZAT_API_KEY"],
)

completion = client.chat.completions.create(
    model="anthropic/claude-sonnet-5-5",
    max_tokens=64,
    messages=[{"role": "user", "content": "Say hello in five words."}],
)
print(completion.choices[0].message.content)

See Chat Completions for what the OpenAI-compatible API supports.

Next stepsCopy Anchor Link

For AI agentsCopy Anchor Link

1. Don't sign up, pay or create keys for the user: those happen in the dashboard at https://router.ezat.io, by the user.
2. Ask the user which environment variable holds their key (EZAT_API_KEY by convention). Never ask them to paste the key into the chat, and never print it.
3. Check the key with GET https://api.router.ezat.io/v1/models (header x-api-key: $EZAT_API_KEY). 200 means it works; 401 means it's missing, mistyped or revoked.
4. Send one small request (max_tokens of 64 or less) to POST https://api.router.ezat.io/v1/messages with a model id from the list. A 402 means the organization has no credit: tell the user to add credit on the Billing page and stop; don't retry.
5. Use https://api.router.ezat.io for Anthropic-format clients and https://api.router.ezat.io/openai/v1 for OpenAI-format clients. Don't add /v1 to the first one.