What Is the Gateway?
Creor Gateway is the default LLM provider built into every Creor installation. It routes your requests through Creor's infrastructure to the underlying model provider (Anthropic, OpenAI, Google, etc.) so you never need to manage individual API keys.
When you first open Creor and sign in with your Creor account, the gateway is ready to use immediately. No additional setup is needed.
How It Works
- You send a message in Creor.
- The Creor engine forwards your request to the gateway API with your Creor auth token.
- The gateway routes the request to the appropriate model provider (e.g., Anthropic for Claude, Google for Gemini).
- The response streams back through the gateway to your editor in real time.
- Token usage is tracked against your Creor account credits.
Note
Supported Models
The gateway model catalog is managed server-side and refreshes automatically every 15 minutes. Available models depend on your Creor plan. Below are commonly available models:
| Model | Provider | Best For |
|---|---|---|
| Claude 4 Sonnet | Anthropic | General coding, complex reasoning |
| Claude 4 Opus | Anthropic | Hardest problems, deep analysis |
| Claude 3.5 Haiku | Anthropic | Fast responses, simple tasks |
| GPT-4o | OpenAI | Broad knowledge, fast |
| o3 | OpenAI | Advanced reasoning |
| o3-mini | OpenAI | Fast reasoning, cost-effective |
| o4-mini | OpenAI | Newest reasoning model |
| Gemini 2.5 Pro | Long context, multimodal | |
| Gemini 2.5 Flash | Fast, cost-effective | |
| Gemini 3 Flash Preview | Cutting-edge speed | |
| Grok 3 | xAI | Reasoning, real-time knowledge |
| Grok 3 Mini | xAI | Fast reasoning |
Tip
Credit-Based Billing
Gateway usage is billed per token through Creor credits. Every Creor account receives credits based on its plan tier. Token costs vary by model -- premium models like Claude 4 Opus and o3 cost more per token than lighter models like Gemini Flash.
How Credits Work
- Credits are denominated in USD. $1 in credits equals $1 of token usage at the provider's rate.
- Input tokens (your prompts, context) and output tokens (model responses) are billed separately.
- You can monitor your usage and remaining credits in the Creor Dashboard.
- When credits run out, gateway requests will fail. Add more credits or switch to a direct API key provider.
For detailed pricing per model, visit the Billing & Credits page in the Dashboard docs.
Getting Started
The gateway is enabled by default. To start using it:
- Sign in to Creor with your Creor account (or create one at creor.ai).
- Open any project in Creor.
- Start chatting -- the gateway is already configured as your default provider.
If you want to explicitly select a gateway model, reference it with the creor-gateway/ prefix:
Configuration
While the gateway works out of the box, you can customize its behavior in creor.json:
Warning
When to Use the Gateway
Ideal For
- Getting started quickly -- no API keys to manage.
- Trying out different models without signing up for multiple provider accounts.
- Teams that want centralized billing through Creor credits.
- Users who prefer a managed experience over self-service API keys.
When to Use Direct API Keys Instead
- You have existing API key agreements with volume discounts.
- You need models that are not yet available on the gateway.
- Your organization requires requests to go directly to the provider (compliance).
- You want to avoid the small gateway routing overhead.