You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've been evaluating your product for use as an LLM Proxy.
It is really difficult to see from the documentation what the client interface with the AI Gateway is. Is it OpenAI API or is it not? Statements such as "The request and response formats are loosely based on OpenAI" are not helpful when there is no documentation explicitly specifying what the interface is and what client SDKs are available.
What did you expect to happen?
I expect to be able to understand very quickly how my application will interact with the AI Gateway.
Code of Conduct and Community Expectations
I agree to follow this project's Code of Conduct
I agree to abide by the Community Expectations
The text was updated successfully, but these errors were encountered:
We can say yes, the llm/v1/completions and llm/v1/chat route types are only valid for part of the OpenAI API format right now.
Statements such as "The request and response formats are loosely based on OpenAI" are not helpful when there is no documentation explicitly specifying what the interface is and what client SDKs are available.
Where is the problem?
https://docs.konghq.com/gateway/latest/get-started/ai-gateway/
What happened?
I've been evaluating your product for use as an LLM Proxy.
It is really difficult to see from the documentation what the client interface with the AI Gateway is. Is it OpenAI API or is it not? Statements such as "The request and response formats are loosely based on OpenAI" are not helpful when there is no documentation explicitly specifying what the interface is and what client SDKs are available.
What did you expect to happen?
I expect to be able to understand very quickly how my application will interact with the AI Gateway.
Code of Conduct and Community Expectations
The text was updated successfully, but these errors were encountered: