Frequently Asked Questions
Basic Questions
Q: What is the maximum timeout for API requests?
A: By default, the API gateway timeout is 30 seconds. For long-running tasks, it's recommended to use asynchronous processing mode.
Q: Does it support batch requests?
A: The current API is designed for single request mode. For batch operations, you can implement concurrent request management on the client side.
Q: How do I get an API Key?
A: API Keys can be applied for and managed in the "API Keys" section of the console.
Technical Questions
Q: How to handle large text input?
A: For large texts exceeding 10MB, it's recommended to process them in chunks first, then merge the results.
Q: Which programming languages are supported?
A: We provide a Python client library. Other languages can be integrated through standard HTTP requests.
Q: How to handle unstable network conditions?
A: It's recommended to implement exponential backoff retry mechanisms with reasonable timeout and retry counts.
Billing Questions
Q: How does the billing work?
A: We use a token-based billing model, including comprehensive calculation of input tokens and output tokens.
Q: Is there a free quota?
A: New users receive a certain amount of free quota upon registration. Please check the console for specific details.
Q: How to view usage statistics?
A: You can view detailed call counts and token usage in the "Usage Statistics" page of the console.