Thanks for the welcome! Yeah, we are hoping to give users some “unavailable” message if the model is currently not available. We could call the completion endpoint and see if it returns any error, but we are hoping to be able to check the health of the model without even calling the completion API.
Here is ChatGPT giving a “rubber ducky” explanation:
APIs, or application programming interfaces, are a set of rules and protocols that allow different software applications to communicate with each other. Think of it like a group of friends trying to play a game together. Each friend has their own set of rules and ways of playing the game, but when they come together, they need to agree on a common set of rules so they can all play together seamlessly. APIs work in a similar way, allowing different software applications to share data and functionality with each other.
If you have questions beyond that, feel free to ask more specifically!