The extraction pipeline accesses LLMs through a LiteLLM proxy with an OpenAI-compatible interface so that model routing is handled through a standardized proxy layer.

LLM access:

— PYRANA_PLATFORM_ARCHITECTURE_REFERENCE.md

Extraction pipeline uses a LiteLLM proxy

— PYRANA_PLATFORM_ARCHITECTURE_REFERENCE.md

(OpenAI-compatible) for model routing.

— PYRANA_PLATFORM_ARCHITECTURE_REFERENCE.md