Austin Schaefer 2b61c80143 Default LLM scoring stage to a model already pulled locally ... Was hardcoded to llama3.1, which isn't actually on this machine. Switched to gemma4-e4b (5.0 GB, already pulled) as a speed/quality balance for a call made once per shortlisted article on every pipeline run.		2026-08-20 17:11:33 +02:00
..
server	Scaffold feedsignal: Rust workspace for LLM-filtered RSS reader	2026-08-20 17:04:14 +02:00
app.rs	Scaffold feedsignal: Rust workspace for LLM-filtered RSS reader	2026-08-20 17:04:14 +02:00
main.rs	Scaffold feedsignal: Rust workspace for LLM-filtered RSS reader	2026-08-20 17:04:14 +02:00
server.rs	Default LLM scoring stage to a model already pulled locally	2026-08-20 17:11:33 +02:00