Originally Xef wanted to implement a proxy server implementation over OpenAI, such that we could rely on the OpenAI OpenAPI Specification to communicate with many different LLM services.
Over the last 18 months, we've learned that the AI market is very fast paced, and as such more powerful tools have been released in the OSS community, that are much more fully featured than we can invest our time into this project.
With keeping costs, and maintenance in mind, we propose to no longer support Xef Proxy Server, and instead recommend using 3rd proxy server such as litellm.
Since we deemed this feature not high qualitative enough, we decided to stop supporting, and maintaining this functionality. If this functionality is important to you, please feel free to copy, or fork the code under the Apache 2.0 LICENSE.
Originally Xef wanted to implement a proxy server implementation over OpenAI, such that we could rely on the OpenAI OpenAPI Specification to communicate with many different LLM services.
Over the last 18 months, we've learned that the AI market is very fast paced, and as such more powerful tools have been released in the OSS community, that are much more fully featured than we can invest our time into this project.
With keeping costs, and maintenance in mind, we propose to no longer support Xef Proxy Server, and instead recommend using 3rd proxy server such as litellm.
Since we deemed this feature not high qualitative enough, we decided to stop supporting, and maintaining this functionality. If this functionality is important to you, please feel free to copy, or fork the code under the Apache 2.0 LICENSE.