DDTokens

English

中文
Русский

English

中文
Русский

Appearance

Frequently Asked Questions ​

What Is DDTokens? ​

DDTokens is an OpenAI-compatible AI model aggregation and distribution gateway. You can use a DDTokens API key to call multiple models and view usage, costs, and logs in the dashboard.

Is It Compatible with the OpenAI SDK? ​

Yes. Most OpenAI SDK projects only need a different API key and base_url:

python
from openai import OpenAI

client = OpenAI(
    api_key="sk-your-key-here",
    base_url="https://apiddt.com/v1",
)

Which Models Are Supported? ​

The dashboard and pricing page are the source of truth for supported models. Common families include OpenAI, Claude, and Gemini. Specific model names, prices, and availability may change with upstream resources.

How Do I Create an API Key? ​

Sign in to the dashboard and create one on the token or API key management page. Use separate keys for production, testing, and local development. See Create API Key for details.

How Am I Charged? ​

DDTokens charges for actual token usage. Cost is generally determined by input tokens, cached tokens, output tokens, the standard model price, and your account discount. See Model Pricing.

Why Can the Same Model Have Different Prices? ​

The same model may be supplied by different providers. Costs, capacity, reliability, and parameter compatibility can vary by provider. Default routing is appropriate for most use cases; read Providers & Routing when you need to pin a provider.

Can I Use Free Models in Production? ​

It is not recommended. Free models are intended for integration tests and lightweight validation and usually have stricter RPM, RPD, and concurrency limits. Production services should use paid models with retries, rate limiting, and logging.

What Should I Check First When a Request Fails? ​

Check in this order:

  1. The API key is valid.
  2. The Base URL is https://apiddt.com/v1.
  3. The account balance and token quota are sufficient.
  4. The model name is available.
  5. The dashboard log contains a request ID and error details.

See Troubleshooting for more steps.

How Can I Control Costs? ​

Create a separate token with its own quota for each environment. Production systems should also record request IDs, models, token usage, and costs for reconciliation and alerts.

How Do I Get Help? ​

If the dashboard provides a ticket system, submit the issue there first. Include the request time, model, status code, error response, and request ID, but never the complete API key.

© 2026 DDTokens