mirror of
https://github.com/arsvendg/Stirling-PDF.git
synced 2026-07-14 18:44:05 +02:00
James BruntonandGitHub
e10c5f6283
# Description of Changes Redesign the Python AI engine to be properly agentic and make use of `pydantic-ai` instead of `langchain` for correctness and ergonomics. This should be a good foundation for us to build our AI engine on going forwards.
10 lines
465 B
Bash
10 lines
465 B
Bash
# Configure the model strings passed to pydantic-ai. Provider credentials are handled by
|
|
# pydantic-ai and should be set using the provider's native environment variables, for example
|
|
# ANTHROPIC_API_KEY or OPENAI_API_KEY.
|
|
STIRLING_SMART_MODEL=anthropic:claude-haiku-4-5
|
|
STIRLING_FAST_MODEL=anthropic:claude-haiku-4-5
|
|
|
|
# Default output token limits applied by the engine for each model tier.
|
|
STIRLING_SMART_MODEL_MAX_TOKENS=8192
|
|
STIRLING_FAST_MODEL_MAX_TOKENS=2048
|