火速AI 文档

English

简体中文

English

简体中文

Appearance

FAQ ​

1. What is HuosuAI? ​

HuosuAI is an intelligent model aggregation platform that provides a unified API gateway to access 40+ leading AI models including OpenAI, Claude, Gemini, DeepSeek, and more. Our API is fully compatible with the OpenAI format, enabling seamless migration for developers. HuosuAI is a product of HanghangAI (hanghang.com).

2. How do I migrate from OpenAI? ​

Simply change two settings: replace the Base URL from https://api.openai.com/v1 to https://api.huosu.com/v1, and use your HuosuAI API Key. No other code changes required — we support all OpenAI features including streaming, function calling, and JSON Mode.

3. Which SDKs are supported? ​

HuosuAI works with any SDK or tool that supports the OpenAI API, including the official OpenAI Python and Node.js SDKs, LangChain, LlamaIndex, and third-party clients like ChatGPT-Next-Web and LobeChat.

4. What if my API Key is compromised? ​

Log in to the console, disable or delete the compromised key immediately, and create a new one. We recommend using environment variables and rotating keys regularly.

5. Why am I getting 429 errors? ​

HTTP 429 means you've exceeded the rate limit (default: 60 requests per minute). Add retry logic with exponential backoff, or contact support to increase your limit.

6. Do I need different API Keys for different models? ​

No. One API Key works for all models. Simply change the model parameter in your request to switch between GPT-4o, Claude, DeepSeek, or any other supported model.

7. Does HuosuAI store my conversation data? ​

No. We do not persistently store conversation content. Request data is only held in memory during processing and cleared immediately after. We only log metadata (timestamps, model used, token counts) for billing purposes.

8. How do I monitor my usage and costs? ​

The HuosuAI console provides a real-time dashboard with usage analytics, detailed call logs filterable by time/model/key, and budget alert settings.

9. Is streaming billed differently? ​

No. Streaming uses the same token-based billing as non-streaming requests. It only changes how responses are delivered (incrementally) without affecting cost.

10. How do I contact support? ​

Intelligent Model Aggregation Platform

© 2026 HuosuAI (huosu.com) | A product of HanghangAI (hanghang.com)