Can you help me understand env variables more

Archon
1

I am setting everything up and would love to know more about what each of these do so i know what to use. and can i use groq for any of them? please help! i don’t want to get this wrong. i am going to build an agent to enrich data, lots of rows of data, i need to be mindful of costs.

  1. Configure the following essential variables:
  • BASE_URL: API endpoint (OpenAI, OpenRouter, or Ollama)
  • LLM_API_KEY: Your API key for the LLM service
  • OPENAI_API_KEY: Required for embeddings
  • SUPABASE_URL: Your Supabase project URL
  • SUPABASE_SERVICE_KEY: Your Supabase service key
  • PRIMARY_MODEL: Main agent model (e.g., gpt-4o-mini)
  • REASONER_MODEL: Planning model (e.g., o3-mini)
1 Like
2

here is my attempt to answer your question :

LLM

  • BASE_URL: API endpoint (OpenAI, OpenRouter, or Ollama)
    BASE_URL is to indiquate where is the llm API. here is the one for openAI :https://api.openai.com/v1
    You should have one in your llm account API settings

  • LLM_API_KEY: Your API key for the LLM service
    Those llm usually requires authentication.

  • OPENAI_API_KEY: Required for embeddings
    embeddings is used to store the chunks into the supabase. I am not sure why it has to be OPENAI.

Supabase

Get your SUPABASE_URL from the API section of your Supabase project settings -
https://supabase.com/dashboard/project//settings/api

  • SUPABASE_URL: Your Supabase project URL
  • SUPABASE_SERVICE_KEY: Your Supabase service key

Archon reasonning agent

You define here which model you will use for chating (PRIMARY) and which one you will use for reasonning. Usually those are differents models.

  • PRIMARY_MODEL: Main agent model (e.g., gpt-4o-mini)
  • REASONER_MODEL: Planning model (e.g., o3-mini)
1 Like
3

Great answer, thank you @renaud!