[Feature]: vLLM support

[fpaupier]

The Feature

(vLLM)[https://github.com/vllm-project/vllm] is a LLM serving framework that enables you to expose openAI API compatible endpoint. How can I get helicone working with vLLM?

Motivation, pitch

For on prem deployment, vLLM is a great option for secure inference data handling and for full control on the model you use. Given helicone can be self deployed with docker compose / K8s, it would be a great complimentary service to have on our infra.

Twitter / LinkedIn details

https://www.linkedin.com/in/fpaupier/