Many customers want to use our OpenAI Inference endpoint against OpenAI compatible API's they have written, or Ollama, or Nvidia Triton OpenAI API front end. I had heard that was the intent of this OpenAI inference endpoint, but we do not state it directly. Can we validate this is OK with Search PM and include this?
Co-authored-by: István Zoltán Szabó <istvan.szabo@elastic.co>