Frequently Asked Questions

Go to the Registration page and fill in your email and password. After registering, you will receive a verification email. Once verified, you can log in and start using the platform immediately. The free tier requires no credit card.

On the login page, click "Forgot Password" and enter your registered email address. You will receive a password reset link that expires within 1 hour. Follow the link to set a new password.

Yes. Go to Settings in your dashboard to update your email address. You will need to verify the new email before the change takes effect.

Contact support via email at bradyliuy@gmail.com to request account deletion. Please note that this action is irreversible and any remaining balance will be forfeited.

Log in to your dashboard, navigate to API Keys, and click "Create Key". Give it a name and optionally set a spending limit. Copy the key immediately — it will only be shown once.

API keys use the format sk- followed by 48 random alphanumeric characters (e.g. sk-a1b2c3d4e5f6...). They are hashed with SHA-256 before storage, so we cannot recover a lost key.

Yes. When creating or editing an API key, you can set a total quota and remaining quota. Once the remaining quota is exhausted, requests using that key will be rejected.

Immediately delete the compromised key from the dashboard and create a new one. Check your usage history for any unauthorized activity. You can also contact support for assistance.

HuiLink uses a pay-as-you-go model. You add funds to your account, and usage is deducted in real-time as requests are processed. There are no monthly commitments or minimums.

We support multiple payment methods: Credit Card (Visa/Mastercard, global), Epay (易支付, Alipay & WeChat Pay for Chinese users), and WeChat Pay Direct (微信支付直连). Choose the method that works best for you in the billing section.

Refunds are handled on a case-by-case basis. Please contact our support team at bradyliuy@gmail.com with your user ID and transaction details. We aim to respond within 1 business day.

Go to the Usage page in your dashboard to see a detailed breakdown of your API usage, including tokens consumed, requests made, and costs incurred. You can filter by date range, model, and API key.

We support 50+ models from 20+ providers including OpenAI (GPT-4o, GPT-4o-mini), DeepSeek (V3, R1), Qwen (Plus, Max), Anthropic (Claude), Google Gemini, and many more. Browse the full list at the Models page.

Models are available through the standard API endpoints. Simply specify the model ID in your request (e.g. "gpt-4o", "deepseek-v4-flash"). If the model is available on your plan, it will work out of the box.

Custom models are not directly supported. However, if you have a specific model you need, please contact us and we can discuss adding it to our platform.

Context length refers to the maximum number of tokens (input + output) a model can process in a single request. For example, a 128K context model can handle approximately 128,000 tokens total per request.

The base URL for all API requests is the gateway URL. Use it as the endpoint for OpenAI-compatible clients. You can find your specific base URL in the API Keys section of your dashboard.

When you exceed the rate limit, the API returns a 429 (Too Many Requests) status code. Rate limits vary by plan: Free (20 RPM), Pay-as-you-go (500 RPM), and Enterprise (custom). Wait and retry after the rate limit window resets.

Errors are returned as standard HTTP status codes with a JSON body containing a code and message. Common errors: 400 (bad request), 401 (auth failure), 402 (insufficient balance), 429 (rate limited), 500 (server error). See the API Reference for the full list.

Yes. Set stream: true in your request body to receive responses via Server-Sent Events (SSE). This is supported for all chat completion models. See the Examples page for code samples in curl, Python, and Node.js.