FAQ

Answers to frequently asked questions

Basic questions

What is the difference between Cova Nexus and using the OpenAI API directly?

Cova Nexus is a unified gateway that, on the basis of being compatible with the OpenAI format, additionally provides: multi-vendor unified access, unified billing, load balancing, failover, usage monitoring and other value-added capabilities. You only need one Token to use multiple models such as Anthropic, OpenAI, Google, and xAI at the same time.

Do I need to register an account with each supplier?

No need. Cova Nexus has established cooperation with various suppliers. You only need to register an account on our platform and create a Token to use all online models.

What programming languages are supported?

Any programming language that supports HTTP requests will do. It is recommended to use OpenAI official SDK (Python, Node.js, Go, etc.), and you only need to modify the base_url to access.

Billing related

How is it billed?

Billing is based on actual Token consumption, and different models have different unit prices. Unit price = model base price × grouping rate. For specific prices, please refer to the "Pricing" page or the labels of each model in the model square.

technical issues

What should I do if the request times out?

It is recommended to set a reasonable timeout (usually 60-120 seconds). If timeouts occur frequently, you can check: whether the model is under high load, whether the request parameters are reasonable (such as max_tokens is too large), and whether the network connection is normal.

How to use streaming output?

Setting "stream": true in the request enables SSE (Server-Sent Events) streaming returns. Fully compatible with standard OpenAI streaming formats.