Your first request

ArkaFlow speaks the OpenAI and Anthropic wire formats. Point your existing SDK at our base URL, keep your code unchanged, and swap models freely.

Base URL

All requests go through the unified gateway:

https://gateway.oya.moe/v1

Send a chat completion

curl

# Any OpenAI-compatible client works unchanged
curl https://gateway.oya.moe/v1/chat/completions \
  -H "Authorization: Bearer $ARKAFLOW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-4-7",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Python

# pip install openai
from openai import OpenAI

client = OpenAI(
    base_url="https://gateway.oya.moe/v1",
    api_key="sk-arka-…",
)

resp = client.chat.completions.create(
    model="claude-opus-4-7",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)

Node.js

// npm install openai
import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://gateway.oya.moe/v1',
  apiKey: process.env.ARKAFLOW_API_KEY,
});

const resp = await client.chat.completions.create({
  model: 'claude-opus-4-7',
  messages: [{role: 'user', content: 'Hello!'}],
});
console.log(resp.choices[0].message.content);

Model routing

The model field accepts any catalog id — the gateway picks the best upstream and fails over automatically. Browse the model catalog for live pricing.

Prefer the Anthropic format?

The gateway also exposes the Anthropic Messages surface — use the official anthropic SDK with the same base URL and key, and call any catalog model through /v1/messages.

Read the response

Responses are byte-compatible with the provider format you called. Usage and cost accounting appear in your console Usage dashboard within seconds.

Next: Go multi-model →