LLM running on Azure using hugging face

I am looking for someone to help me deploy a metagpt from GitHub on Azure and using litellm/proxyllm switch it from using openai to Llama2 70B on huggingface with a tie in to be connected to my GitHub. I need clear step by step with image documentation on how it was set up, configs that can be change, etc. Ideally I want to deploy chatdev on Azure and using litellm/proxyllm/LMStudio (not particular), change it to using llama2 70b on huggingface, and also integrate gpt engineer as a role to help the CEO role define the project (using llama2 on hugging face) and gpt researcher to find potential solutions that could be leveraged and code that could be used off of GitHub (using llama2 on hugging face) as a role. I don’t want openai used. Needs to be internet accessible. I want any code used or created to be put into my private GitHub and to use a private huggingface space.