Add a working path to call a commercial model API alongside the local Ollama option,
with configuration and documentation.
The local option has to stay available. Some data cannot be sent to an outside service,
and pinning an exact local model version is what makes a run reproducible.
Include: configuration for API keys, sensible error handling when a call fails, and clear
documentation of what leaves the machine on each path.
Add a working path to call a commercial model API alongside the local Ollama option,
with configuration and documentation.
The local option has to stay available. Some data cannot be sent to an outside service,
and pinning an exact local model version is what makes a run reproducible.
Include: configuration for API keys, sensible error handling when a call fails, and clear
documentation of what leaves the machine on each path.