Quickstart

From signing in to a working integration. Each step takes a minute or two, and you can stop after step 3 if all you need is one call.

1. Get your API Key

Sign in and navigate to Console → API Keys. Click "Create key" and copy it immediately - it will only be shown once.

2. Make your first request

Pick the SDK you already use. Both recipes hit the same gateway with the same API key.

Option A - OpenAI SDK

curl https://synthorai.io/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gpt-5.4-mini",
    "messages": [
      {"role": "user", "content": "Hello! What can you do?"}
    ]
  }'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://synthorai.io/v1"
)

response = client.chat.completions.create(
    model="gpt-5.4-mini",
    messages=[{"role": "user", "content": "Hello! What can you do?"}]
)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://synthorai.io/v1",
});

const response = await client.chat.completions.create({
  model: "gpt-5.4-mini",
  messages: [{ role: "user", content: "Hello! What can you do?" }],
});

console.log(response.choices[0].message.content);

Option B - Anthropic SDK

curl https://synthorai.io/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "claude-sonnet-4-6",
    "max_tokens": 256,
    "messages": [
      {"role": "user", "content": "Hello! What can you do?"}
    ]
  }'

from anthropic import Anthropic

client = Anthropic(
    api_key="YOUR_API_KEY",
    base_url="https://synthorai.io"
)

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=256,
    messages=[{"role": "user", "content": "Hello! What can you do?"}]
)

print(message.content[0].text)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://synthorai.io",
});

const message = await client.messages.create({
  model: "claude-sonnet-4-6",
  max_tokens: 256,
  messages: [{ role: "user", content: "Hello! What can you do?" }],
});

console.log(message.content[0].text);

3. Try it without writing code

Open the Playground in the console, pick a model and send a prompt. It runs through the same pipeline as your API traffic, so a reply confirms the model and your balance. It authenticates with your console session rather than an API key, so test the key itself with the request above.

• Playground →

4. List available models

curl https://synthorai.io/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"

5. Wire it into your app

Point your existing client at the gateway: change the base URL and use your Synthorai key. Your model ids, request shapes and SDKs stay as they are. From there the capability guides cover the things that change your bill - caching a repeated prefix, or capping how much a model thinks.

Using Claude Code, Codex CLI or OpenClaw instead of an SDK? Those have their own one-line setup:

• Claude Code →
• OpenAI Codex CLI →
• OpenClaw →
• Tool Compatibility →

6. Add credit when you need it

Usage draws down a prepaid balance - no monthly platform fee, no subscription. Top up from Billing with a card or crypto; the balance is usable the moment it lands. Free credit comes from a welcome campaign where one is running, not automatically on signup.

• Billing & Top-up →

Where to go next

• Prompt caching - bill a repeated prefix at a fraction of its normal price
• Reasoning effort - the biggest cost lever on models that reason by default
• API keys - quotas, model allowlists, expiry and rotation
• Usage analytics - where the spend actually went

Authentication

All API requests require an API key passed in the Authorization header.

Bearer Token

Authorization: Bearer YOUR_API_KEY

Key Scopes

Key Format

Synthorai keys are prefixed with sk- followed by a random 32-character string. Keys are hashed at rest and cannot be recovered after creation.

Not sure which model to start with? Browse the model price comparison to see every routable model's per-token pricing side by side.