OpeniBank logoOpeniBank
Docs

Local LLM

Run OpeniBank with local or cloud LLMs using simple environment variables.

OpeniBank is local-first by default, but you can mix local and cloud models depending on the workload.

Run with Ollama

  1. Install Ollama: https://ollama.com
  2. Pull a model:
ollama pull llama3.1
  1. Point OpeniBank to the local provider:
export OPENIBANK_LLM_PROVIDER=ollama
export OPENIBANK_LLM_MODEL=llama3.1
export OPENIBANK_LLM_BASE_URL=http://localhost:11434

Run with cloud APIs

export OPENIBANK_LLM_PROVIDER=anthropic
export OPENIBANK_LLM_MODEL=claude-3-5-sonnet
export OPENIBANK_LLM_API_KEY=your_api_key

export OPENIBANK_LLM_PROVIDER=openai
export OPENIBANK_LLM_MODEL=gpt-4o-mini
export OPENIBANK_LLM_API_KEY=your_api_key

Resonators still enforce commitments, permits, and receipts regardless of the LLM provider.